Data analysis method and device, electronic equipment, storage medium and program product

Abstract

The embodiment of the invention provides a data analysis method and device, electronic equipment, a storage medium and a program product. The method comprises the steps that data analysis chapters matched with received data analysis requirements are selected from a chapter library, chapter information of the data analysis chapters is obtained, and the chapter information comprises chapter analysis attribute information and a chapter generation strategy set; generating a data analysis content framework based on the chapter analysis attribute information of the data analysis chapter and the chapter generation strategy set; generating chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation strategy set of the data analysis chapter and the data analysis requirement; and splicing the chapter content of the data analysis chapter to obtain data analysis content meeting the data analysis requirement. By means of the technical scheme, the generation efficiency and accuracy of the data analysis content can be improved.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technologies, and in particular, to a data analysis method, apparatus, electronic device, storage medium, and program product. Background Art

[0002] During the business operation process of an enterprise, a large amount of business data will be accumulated. In-depth analysis of the business data from different scenarios and perspectives helps the enterprise better manage its business.

[0003] In related technologies, data mining and analysis mainly rely on the experience, knowledge of data analysis personnel, and the quality of business data. The entire process of data mining and analysis has a long cycle, high input costs, and poor accuracy of the generated data analysis content, and requires repeated communication and calibration. Summary of the Invention

[0004] Embodiments of the present disclosure provide a data analysis method, apparatus, electronic device, storage medium, and program product to improve the generation efficiency and accuracy of data analysis content.

[0005] In a first aspect, embodiments of the present disclosure provide a data analysis method, including:

[0006] Selecting a data analysis chapter that matches the received data analysis requirement from a chapter library, and obtaining chapter information of the data analysis chapter, where the chapter information includes chapter analysis attribute information and a chapter generation strategy set;

[0007] Generating a data analysis content framework based on the chapter analysis attribute information and the chapter generation strategy set of the data analysis chapter;

[0008] Generating chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation strategy set of the data analysis chapter, and the data analysis requirement;

[0009] Concatenating the chapter content of the data analysis chapter to obtain data analysis content that meets the data analysis requirement.

[0010] In a second aspect, embodiments of the present disclosure further provide a data analysis apparatus, including:

[0011] An information acquisition module, configured to select a data analysis chapter that matches the received data analysis requirement from a chapter library, and obtain chapter information of the data analysis chapter, where the chapter information includes chapter analysis attribute information and a chapter generation strategy set;

[0012] A framework generation module, configured to generate a data analysis content framework based on the chapter analysis attribute information and the chapter generation strategy set of the data analysis chapter;

[0013] A chapter generation module, configured to generate the chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation strategy set of the data analysis chapter, and the data analysis requirements;

[0014] A chapter splicing module, configured to splice the chapter content of the data analysis chapter to obtain data analysis content that meets the data analysis requirements.

[0015] Thirdly, an embodiment of the present disclosure further provides an electronic device, including:

[0016] One or more processors;

[0017] A memory, configured to store one or more programs,

[0018] When the one or more programs are executed by the one or more processors, the one or more processors implement the data analysis method as described in the embodiments of the present disclosure.

[0019] Fourthly, an embodiment of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the data analysis method as described in the embodiments of the present disclosure.

[0020] Fifthly, an embodiment of the present disclosure further provides a computer program product, when the computer program product is executed by a computer, the computer implements the data analysis method as described in the embodiments of the present disclosure.

[0021] The data analysis method, device, electronic device, storage medium, and program product provided by the embodiments of the present disclosure pre-configure a chapter library and configure chapter generation strategies for each supported chapter in the chapter library. After receiving data analysis requirements, a data analysis chapter that matches the received data analysis requirements is selected from this chapter library to generate a data analysis content framework, and the chapter content of each data analysis chapter in the data analysis content framework is generated according to the chapter generation strategies of each data analysis chapter, which can automatically generate data analysis content that meets the data analysis requirements based on the pre-configured strategies. Compared with the technical solutions of manually performing data mining and analysis or completely relying on large language models to generate data analysis content, it can improve the generation efficiency and accuracy of data analysis content. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In combination with the accompanying drawings and with reference to the following specific embodiments, the above and other features, advantages, and aspects of the embodiments of the present disclosure will become more obvious. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic, and the original elements and elements are not necessarily drawn to scale.

[0023] Figure 1Flow diagram of a data analysis method provided by an embodiment of the present disclosure;

[0024] Figure 2 Flow diagram of another data analysis method provided by an embodiment of the present disclosure;

[0025] Figure 3 Schematic diagram of a data analysis process provided by an embodiment of the present disclosure;

[0026] Figure 4 Block diagram of the structure of a data analysis device provided by an embodiment of the present disclosure;

[0027] Figure 5 Schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure. Detailed implementation manners

[0028] Hereinafter, embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not used to limit the protection scope of the present disclosure.

[0029] It should be understood that the various steps recited in the method embodiments of the present disclosure can be executed in a different order and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this regard.

[0030] As used herein, the term "including" and its variations are open-ended, that is, "including but not limited to". The term "based on" is "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0031] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependent relationships.

[0032] It should be noted that the modifications of "one" and "multiple" mentioned in the present disclosure are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly specified in the context, it should be understood as "one or more".

[0033] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are for illustrative purposes only and are not used to limit the scope of these messages or information.

[0034] It can be understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) should comply with the requirements of the corresponding laws, regulations and related provisions.

[0035] Figure 1 The following is a schematic flowchart of a data analysis method provided by an embodiment of the present disclosure. This method can be executed by a data analysis device, where the device can be implemented by software and / or hardware and can be configured in an electronic device. Typically, it can be configured in a computer, a mobile phone or a tablet computer. The data analysis method provided by the embodiments of the present disclosure is applicable to scenarios where data analysis content is generated based on the data analysis requirements input by a user. As Figure 1 shown, the data analysis method provided in this embodiment may include:

[0036] S101. Select a data analysis chapter that matches the received data analysis requirements from the chapter library, and obtain the chapter information of the data analysis chapter, where the chapter information includes chapter analysis attribute information and a chapter generation strategy set.

[0037] The data analysis requirements can be understood as the requirements for the data analysis content to be generated, and it can be the original description of the data analysis to be performed. These data analysis requirements can be input by a user or sent by other devices.

[0038] The chapter library can be understood as a pre-set set of candidate chapters, where the chapters can include content formats such as text, pictures, voices, videos, etc. The chapters can also be documents or content paragraphs in a document. The chapter library can be pre-configured, and this chapter library can include the generation strategy configuration of the chapters that support generation. For example, the chapter library can include multiple data analysis chapters and the chapter generation strategy set of each data analysis chapter. The chapter generation strategy set of each data analysis chapter can include one or more chapter generation strategies for this data analysis chapter.

[0039] The chapter generation strategy can be the generation strategy of the chapter content, and it can be a description of the generation method of the chapter content. The type of the chapter generation strategy is not limited. For example, the chapter generation strategy can include but is not limited to the parameter extraction strategy and / or the data analysis strategy of the corresponding data analysis chapter. Optionally, the chapter generation strategy set includes a parameter extraction strategy subset and a data analysis strategy subset, where the parameter extraction strategy subset includes at least one parameter extraction strategy; the data analysis strategy subset includes a first data analysis strategy based on a strategy template and / or a second data analysis strategy not based on a strategy template.

[0040] Among them, the parameter extraction strategy subset can be understood as the set of parameter extraction strategies for the corresponding data analysis section. The parameter extraction strategy for a certain data analysis section can be used to describe the data analysis parameter extraction method for this data analysis section. For example, it can be used to describe the parameters required in the data analysis process of this data analysis section. These data analysis parameters can be used as the common parameters for this data analysis section in the data analysis process.

[0041] The data analysis strategy subset can be understood as the set of data analysis strategies for the corresponding data analysis section. This data analysis strategy can be used to describe the data analysis method for this data analysis section. The data analysis strategy subset for a certain data analysis section can include one or more data analysis strategies for this data analysis section. For example, it can include at least one first data analysis strategy and / or at least one second data analysis strategy for this data analysis section. The first data analysis strategy and the second data analysis strategy can be different types of data analysis strategies. Exemplarily, the first data analysis strategy can be a data analysis strategy based on a strategy template, such as the data analysis strategy included in a pre-set strategy template. The second data analysis strategy can be a data analysis strategy that does not depend on a pre-set strategy template, such as a data analysis strategy that fully complies with the relevant knowledge of the data analysis section and the user's data analysis requirements.

[0042] The section information of the data analysis section can include the section generation strategy set for this data analysis section, and can further include the section analysis attribute information for this section. This section analysis attribute information can be used to describe the data analysis attributes of the data analysis section. Exemplarily, the analysis attribute information of the data analysis section can include, but is not limited to, the analysis topic attribute information of the data analysis section. This analysis topic attribute information can be used to describe the data analysis topic of the data analysis section.

[0043] Specifically, it is possible to receive data analysis requirements; select one or more data analysis sections that match these data analysis requirements from a pre-set section library; and obtain the section information of each selected data analysis section. For example, obtain the section analysis attribute information and the section generation strategy set of each selected data analysis section.

[0044] When selecting a data analysis section that matches the received data analysis requirements, exemplarily, sections that match the data analysis requirements can be screened from the section library and used as the data analysis sections included in the data analysis content to be generated. For example, the received data analysis requirements are input into a preset model (such as a large language model, etc.) that has been pre-trained, and the understanding and text extraction capabilities of the preset model are used to extract the key data analysis intentions and analysis objectives (such as analysis objects) for this time from the received data analysis requirements; after the key analysis intentions and analysis objectives for this time are extracted, section screening can be performed based on these key analysis intentions and analysis objectives. For example, candidate sections that meet these key analysis intentions and analysis objectives are screened from the section library, and several sections are selected as the data analysis sections that make up the data analysis content in the order of the relevance between the selected candidate sections and these key analysis intentions and analysis objectives from high to low.

[0045] After screening the data analysis sections that match the received data analysis requirements from the section library, the section analysis attribute information and section generation strategy set of each data analysis section can be obtained. For example, the section analysis attribute information and section generation strategy set of each section can be preset and stored in the section library in advance. Thus, the section analysis attribute information and section generation strategy set of the screened data analysis sections can be obtained from the section library.

[0046] S102. Generate a data analysis content framework based on the section analysis attribute information and section generation strategy set of the data analysis section.

[0047] Among them, the data analysis content framework can be understood as the framework structure of the data analysis content to be generated. This framework structure can refer to the overall structure and components of the data analysis content. Exemplarily, the data analysis content framework can include one or more of the content title, content summary of the data analysis content, and section framework information of each data analysis section in the data analysis content. This embodiment does not make any limitations in this regard.

[0048] Specifically, after obtaining the data analysis sections that match the data analysis requirements and the section information of the data analysis sections, a data analysis content framework including the selected data analysis sections can be generated based on the section information of this data analysis section. For example, the section framework information of each data analysis section is spliced according to a preset framework structure to obtain the data analysis content framework of the data analysis content to be generated, so as to facilitate the subsequent generation of the data analysis content based on this data analysis content framework.

[0049] In this embodiment, the data analysis content framework may include the framework information of each selected data analysis chapter. The arrangement order of the framework information of each data analysis chapter in the data analysis content framework is not limited. For example, the arrangement order of the framework information of each data analysis chapter in the data analysis content framework may be determined according to the preset weight values of each data analysis chapter, the relationships between each data analysis chapter, and / or the degree of relevance between each data analysis chapter and the received data analysis requirements, etc.

[0050] In this embodiment, the data analysis content framework may further include the content title and / or content abstract of the data analysis content to be generated. This content title can be understood as the title of the data analysis content to be generated; this content abstract can be understood as the abstract of the data analysis content to be generated. This abstract can be used to summarize the key points of the data analysis content to be generated. These key points may include, but are not limited to, the overall analysis method, analysis idea, and / or relevant background of the data analysis content, etc. This content title and / or content abstract can be generated based on the received data analysis requirements, the obtained chapter analysis attribute information of each data analysis chapter, and / or the framework information of each data analysis chapter, etc. When generating the content title and / or content abstract of the data analysis content, the arrangement order of each data analysis chapter in the data analysis content framework may not be considered to improve the generation rate of the data analysis content framework; or the arrangement order of each data analysis chapter in the data analysis content framework may be considered to improve the accuracy of the generated content title and / or content abstract.

[0051] In some embodiments, generating the data analysis content framework based on the chapter analysis attribute information and chapter generation strategy set of the data analysis chapter includes: generating the framework information of the data analysis chapter according to the chapter analysis attribute information and chapter generation strategy set of the data analysis chapter; sorting the framework information according to the matching degree between the data analysis chapter and the data analysis requirements to obtain the sorting result of the framework information; generating the content title and content abstract of the data analysis content based on the sorting result and the chapter analysis attribute information of the data analysis chapter; splicing the content title, the content abstract, and the framework information of the data analysis chapter to obtain the content framework of the data analysis content.

[0052] Among them, the chapter framework information of a certain data analysis chapter can be understood as the framework information of this data analysis chapter included in the data analysis content framework. In other words, it is the framework content presented by this data analysis chapter in the data analysis content framework. Exemplarily, the chapter framework information of the data analysis chapter can include the chapter title information, analysis scenario information, analysis method information, and analysis conclusion information of the data analysis chapter. The analysis scenario information can be used to describe the analysis scenario of the data analysis chapter, such as the scenario introduction information of the data analysis chapter. The analysis method information can be used to describe the analysis method of the data analysis chapter, such as the overview information of the analysis method, etc. The analysis conclusion information can be used to describe the analysis conclusion of the data analysis chapter or the generation method of the analysis conclusion, etc. The matching degree between the data analysis chapter and the data analysis requirements can be used to characterize the matching degree between the data analysis chapter and the data analysis requirements, such as the correlation degree between the data analysis chapter and the received data sharing requirements. The calculation method of this matching degree is not limited. For example, the matching degree between the data analysis chapter and the received data analysis requirements can be calculated through a pre-set matching degree calculation rule; it can also be calculated through a pre-trained model (such as a large language model or an embedded model, etc.) for the matching degree between the data analysis chapter and the received data analysis requirements, and so on.

[0053] Exemplarily, the chapter framework information of the data analysis framework can be generated based on the chapter analysis attribute information and the chapter generation strategy set of the data analysis chapter. For example, the chapter analysis attribute information and the chapter generation strategy set of each selected data analysis chapter are input into a preset model, and the chapter framework information of each data analysis chapter is generated through this preset model, and so on. Moreover, according to the chapter information such as the chapter analysis attribute information and the chapter generation strategy set of each data analysis chapter, and / or according to the chapter framework information of each data analysis chapter, the correlation degree between each data analysis chapter and the data analysis requirements is calculated as the matching degree between the data analysis chapter and the data analysis requirements. The data analysis chapters are sorted in descending order according to this matching degree to obtain the sorting result of each data analysis chapter. For example, the chapter framework information of each data analysis chapter is sorted in descending order according to this matching degree to obtain the sorting result of the chapter framework information of each data analysis chapter as the sorting result of each data analysis chapter. According to the sorting result of each data analysis chapter and the chapter analysis attribute information (such as the chapter analysis theme, etc.) of each data analysis chapter, the content theme and content summary of the data analysis content are generated. After that, the generated chapter titles, chapter summaries, and the chapter framework information of each data analysis chapter can be concatenated in sequence to obtain the data analysis content framework. Among them, in the data analysis content framework, the content summary can be located after the content title, and the chapter framework information of each data analysis chapter can be located after the content summary. The chapter framework information of each data analysis chapter can be concatenated based on the sorting result of each data analysis chapter to preferentially display the data analysis chapters with higher relevance to the received data analysis requirements in the data analysis content.

[0054] In some examples, after generating the data analysis content framework, the data analysis content framework can also be presented to the user to facilitate the user to adjust or modify the generated data analysis content framework. After the user completes the adjustment or modification of the data analysis content framework, subsequently, the chapter content of each data analysis chapter in the data analysis content framework can be generated based on the adjusted or modified data analysis content framework by the user. For example, after the user completes the confirmation of the data analysis content framework, the chapter content of each data analysis chapter in the data analysis content framework can be triggered to be generated in parallel.

[0055] S103. Generate the chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation strategy set of the data analysis chapter, and the data analysis requirements.

[0056] In this embodiment, after generating the data analysis content framework, for each data analysis chapter in the data analysis content framework, the chapter content of this data analysis chapter can be generated according to this data analysis content framework, the chapter generation strategy set of this data analysis chapter, and the received data analysis requirements.

[0057] Exemplarily, for each data analysis chapter in the data analysis content framework, one or more chapter generation strategies in the chapter generation strategy set of this data analysis chapter can be used to generate chapter content that matches the received data analysis requirements and the chapter framework information of this data analysis chapter in the data analysis content framework. For example, based on one or more chapter generation strategies in the chapter generation strategy set of this data analysis chapter, data that matches the received data analysis requirements and the chapter framework information of this data analysis chapter is obtained, one or more data reports are generated based on this data, and the data reports are spliced together to serve as the chapter content of this data analysis chapter, etc.

[0058] In addition, the data analysis framework may include one or more data analysis chapters. For example, the data analysis objectives of this chapter can be disassembled into problems, and the detailed data breakdown analysis and analysis conclusions of each sub-problem obtained from the disassembly are described through different data analysis chapters. In addition to including one or more data reports, the chapter content of the data analysis chapter may further include a chapter description and / or a chapter summary of this data analysis chapter. This chapter description can be understood as a chapter abstract, which can be used to describe the scenario and / or analysis method of the data analysis sub-problems of this data analysis chapter. The chapter summary can be summary information generated for the analysis scenario corresponding to this data analysis chapter based on the data reports of this data analysis chapter.

[0059] In some examples, after splicing the data reports in this data analysis chapter, the chapter description and chapter summary of this data analysis chapter can be generated based on the spliced content. The generated chapter description is added before the spliced content, and the generated chapter summary is added after the spliced content, thereby obtaining the chapter content of this data analysis chapter.

[0060] S104. Splice the chapter content of the data analysis chapter to obtain data analysis content that meets the data analysis requirements.

[0061] In this embodiment, after generating the chapter content of each data analysis chapter in the data analysis content framework, the chapter content of each data analysis chapter can be spliced. Exemplarily, the chapter content of each data analysis chapter is spliced based on the data analysis content framework. For example, the chapter framework information of the corresponding data analysis chapter in the data analysis content framework except for the chapter title is replaced with the chapter content of each data analysis chapter, thereby obtaining data analysis content that meets the received data analysis requirements.

[0062] In some examples, in addition to the chapter content of each data analysis chapter, the data analysis content may further include a content conclusion. This content conclusion can be a summary of the overall data analysis content. For example, this content summary may include answers to the questions mentioned in the received data analysis requirements, conclusions, and relevant suggestions, etc. In such a case, after splicing the chapter content of each data analysis chapter, a content summary corresponding to the received data analysis requirements can be generated based on the spliced chapter content and added after the spliced chapter content.

[0063] In this embodiment, the presentation form of the generated data analysis content is not limited. Taking the example that the generated data analysis content is presented in the form of document content, exemplarily, the generated data analysis content can be presented as the report content of a data analysis report, etc.

[0064] The data analysis method provided in this embodiment selects data analysis chapters that match the received data analysis requirements from the chapter library and obtains the chapter information of the selected data analysis chapters. This chapter information includes chapter analysis attribute information and a chapter generation strategy set; generates a data analysis content framework based on the chapter analysis attribute information and the chapter generation strategy set of the data analysis chapters; generates the chapter content of each data analysis chapter according to this data analysis content framework, the chapter generation strategy of the data analysis chapters, and the received data analysis requirements; and splices the chapter content of each data analysis chapter to obtain data analysis content that meets the received data analysis requirements. By using the above technical solution, this embodiment pre-configures a chapter library and configures the chapter generation strategies for each supported chapter in the chapter library. After receiving a data analysis requirement, it selects a data analysis chapter that matches the received data analysis requirement from this chapter library to generate a data analysis content framework, and generates the chapter content of each data analysis chapter in the data analysis content framework according to the chapter generation strategy of each data analysis chapter, which can automatically generate data analysis content that meets the data analysis requirements based on the pre-configured strategy. Compared with the technical solutions of manually performing data mining and analysis or completely relying on large language models to generate data analysis content, it can improve the generation efficiency and accuracy of data analysis content.

[0065] Figure 2A flowchart of another data analysis method provided by an embodiment of the present disclosure. The solution in this embodiment can be combined with one or more alternative solutions in the above embodiments. Optionally, generating the chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation strategy set of the data analysis chapter, and the data analysis requirements includes: extracting the data analysis parameters of the data analysis chapter from the data analysis requirements and the data analysis content framework according to the parameter extraction strategy of the data analysis chapter; performing data analysis based on the data analysis parameters and at least part of the data analysis strategies of the data analysis chapter to obtain a data report corresponding to the at least part of the data analysis strategies, where the data report includes report data and a data analysis conclusion; splicing the data reports to obtain the chapter content of the data analysis chapter.

[0066] Correspondingly, as Figure 2 shown, the data analysis method provided in this embodiment may include:

[0067] S201. Select a data analysis chapter that matches the received data analysis requirements from the chapter library, and obtain the chapter information of the data analysis chapter, where the chapter information includes chapter analysis attribute information and a chapter generation strategy set.

[0068] The chapter generation strategy set includes a parameter extraction strategy subset and a data analysis strategy subset. Among them, the parameter extraction strategy subset includes at least one parameter extraction strategy; the data analysis strategy subset includes a first data analysis strategy based on a strategy template and / or a second data analysis strategy not based on a strategy template.

[0069] S202. Generate a data analysis content framework based on the chapter analysis attribute information and the chapter generation strategy set of the data analysis chapter.

[0070] S203. Extract the data analysis parameters of the data analysis chapter from the data analysis requirements and the data analysis content framework according to the parameter extraction strategy of the data analysis chapter.

[0071] Exemplarily, for each data analysis chapter in the data analysis content framework, according to one or more parameter extraction strategies of this data analysis chapter, the common parameters required for this data analysis chapter can be extracted from the received data analysis requirements and the generated data analysis content framework as the data analysis parameters of this data analysis chapter, so as to facilitate generating the chapter content of this data analysis chapter based on this data analysis parameter subsequently.

[0072] S204. Perform data analysis based on at least part of the data analysis strategies of the data analysis parameters and the data analysis chapter to obtain a data report corresponding to the at least part of the data analysis strategies. The data report includes report data and data analysis conclusions.

[0073] In this embodiment, after the data analysis parameters of this data analysis chapter are extracted, data analysis can be performed based on these data analysis parameters and at least part of the data analysis strategies of this data analysis chapter to generate report data and data analysis conclusions corresponding to this at least part of the data analysis strategies, as the data report corresponding to this at least part of the data analysis strategies.

[0074] Exemplarily, one or more data analysis strategies of this data analysis chapter can be obtained from the data analysis strategy subset of the data analysis chapter; for each obtained data analysis strategy, data analysis is performed according to this data analysis strategy and the data analysis parameters of this data analysis chapter. For example, report data corresponding to this data analysis strategy is generated based on this data analysis strategy and data analysis parameters, and data analysis of this report data is performed to generate the data analysis conclusion of this report data, thereby obtaining the data report corresponding to this data analysis strategy.

[0075] In some embodiments, it is possible to support data analysis based on the data analysis strategies (i.e., the first data analysis strategies) in the pre-set strategy templates to improve the generation speed and accuracy of data reports. In this case, optionally, the performing data analysis based on the data analysis parameters and at least part of the data analysis strategies of the data analysis chapter to obtain a data report corresponding to the at least part of the data analysis strategies includes: selecting a pre-set strategy template that matches the data analysis requirements from the strategy template set of the data analysis chapter. The pre-set strategy template includes the pre-set first data analysis strategy and the data extraction code corresponding to the first data analysis strategy; adjusting the pre-set strategy template based on the data analysis parameters and the first knowledge item. The first knowledge item is the knowledge item in the knowledge base of the data analysis chapter corresponding to the pre-set strategy template; using the adjusted pre-set strategy template to perform data analysis to obtain a data report corresponding to the pre-set strategy template.

[0076] Among them, the policy template set can be understood as a collection of policy templates. The policy template set in the data analysis chapter can include one or more pre-set policy templates in this data analysis chapter. Different policy templates can record different pre-set first data analysis policies, and different first data analysis policies can be used to analyze different report data. In addition, the policy template can further record the data extraction code corresponding to the first data analysis policy it contains, and this data extraction code can be used to extract the report data corresponding to this first data analysis policy. The preset policy template can be a policy template selected from the policy template set in the data analysis chapter, and the number can be one or more. The first knowledge item can be the knowledge item in the knowledge base of the data analysis chapter corresponding to the preset policy template, such as the knowledge item that can be used to generate the report data corresponding to this preset policy template. The knowledge base of the data analysis chapter can be used to store the knowledge items related to this data analysis chapter. Such knowledge items can include, for example, documents, charts, and / or documents related to the data analysis chapter.

[0077] Exemplarily, a policy template that meets the received data analysis requirements can be selected from the policy template set in the data analysis chapter as the preset policy template. For example, by using the vectorization method or the ability of the preset model, a preset policy template associated with the data analysis intention and analysis goal of this data analysis chapter determined based on the data analysis requirements can be screened out from the policy template set. After obtaining one or more preset policy templates, for each preset policy template, the knowledge item corresponding to this preset policy template can be obtained from the knowledge base of this data analysis chapter, such as the knowledge item related to this preset policy template and the data analysis intention and analysis goal of this data analysis chapter, as the first knowledge item. Based on the data analysis parameters of this data analysis chapter and the obtained first knowledge item, this preset policy template can be fine-tuned, such as fine-tuning the data extraction code and the first data analysis policy in the preset policy template, so that the preset policy template better meets the actual analysis needs of this data analysis chapter. After the fine-tuning is completed, data analysis can be performed based on the fine-tuned preset policy template to obtain a data report corresponding to this fine-tuned preset policy template.

[0078] In the above embodiment, one or more screening methods can be used to screen the preset policy template that meets the data analysis requirements from the policy template set in the data analysis chapter. Such screening methods can include, for example, screening by vectorization or screening by semantic screening, etc., and this embodiment does not limit this.

[0079] In some embodiments, selecting a preset policy template that matches the data analysis requirements from the policy template set of the data analysis chapter includes: calculating the matching degree between the policy templates in the policy template set and the data analysis requirements by using at least one matching degree calculation method, and selecting at least one policy template from the policy template set according to the matching degree as the preset policy template that matches the data analysis requirements, where the at least one matching method includes a vector matching method and / or a semantic matching method.

[0080] Specifically, the matching degree between each policy template in the policy template set and the received data analysis requirements can be calculated by using a vector matching method and / or a semantic matching method, and one or more policy templates whose matching degree meets the preset conditions can be selected from the policy template set as the preset policy template.

[0081] As an optional example, when the number of policy templates to be screened is small, such as when the number of policy templates included in the policy template set of the data analysis chapter is less than or equal to the first quantity threshold, a semantic matching method can be used to determine the preset policy template. For example, by using the capabilities of a preset model, the report data names, analysis intentions, and objectives recorded in each policy template in the policy template set are judged, and several policy templates that are most relevant to the received data analysis requirements and the data analysis objectives of this data analysis chapter are selected as the preset policy templates.

[0082] When the number of policy templates to be screened is large, such as when the number of policy templates included in the policy template set of the data analysis chapter is greater than the first quantity threshold, a vector matching method can be used to preliminarily screen out at least one candidate policy template. For example, a pre-trained embedding model is used to calculate the first feature vector of the received data analysis requirements and the second feature vector of the data analysis objectives of this data analysis chapter in the data analysis content framework; the first feature vector and the second feature vector are respectively used to perform a correlation calculation with the analysis intentions and objectives of each policy template in the policy template set to obtain at least one candidate policy template whose degree of relevance meets the preset requirements.

[0083] After at least one candidate policy template is preliminarily screened out by using the vector matching method, if the number of the screened candidate policy templates is small, such as the number of the screened candidate policy templates is less than or equal to the second quantity threshold, then this at least one candidate policy template preliminarily screened out can be directly used as the preset policy template; if the number of the screened candidate policy templates is large, such as the number of the screened candidate policy templates is greater than the second quantity threshold, then a semantic matching method can be further used to screen out one or more candidate policy templates from this at least one candidate policy template as the preset policy template.

[0084] In the above embodiments, data aggregation calculation can be performed based on the adjusted preset policy template and corresponding data result analysis can be carried out. For example, data extraction is performed based on the adjusted data extraction code in this preset policy template to generate report data corresponding to this preset policy template, and through the adjusted first data analysis policy in this preset policy template, the generated report data is analyzed to generate a data analysis conclusion of this report data, thereby obtaining a data report corresponding to this preset policy template. In this case, optionally, performing data analysis using the adjusted preset policy template to obtain a data report corresponding to the preset policy template includes: performing data extraction using the adjusted data extraction code in the preset policy template to obtain first report data corresponding to the preset policy template; performing data analysis on the first report data based on the adjusted first data analysis policy in the preset policy template to obtain a first data analysis conclusion of the first report data. Among them, the first report data can be understood as report data generated from the data extracted according to the data extraction code in the preset policy template, and the first data analysis conclusion can be understood as the analysis conclusion of the first report data.

[0085] In some embodiments, it is possible to support data analysis without relying on a pre-set policy template to fully explore the implicit information in the corresponding scenario and improve the comprehensiveness of the content of the data analysis section. In this case, optionally, performing data analysis based on the data analysis parameters and at least part of the data analysis strategies in the data analysis section to obtain a data report corresponding to at least part of the data analysis strategies includes: determining the data analysis logic of the data analysis section according to the data analysis parameters, the second data analysis strategy, and the second knowledge item, where the second knowledge item is the knowledge item corresponding to the second data analysis logic in the knowledge base of the data analysis section; generating a data detail table corresponding to the data analysis logic, where the data detail table at least indicates the associated fields of the data analysis logic; generating a data report corresponding to the second data analysis strategy based on the data detail table.

[0086] Among them, the second knowledge item can be understood as the knowledge item in the knowledge base of the data analysis chapter corresponding to the second data analysis strategy of this data analysis chapter. The data analysis logic can be understood as the analysis logic corresponding to the second data analysis strategy. Exemplarily, this data analysis logic can be used to describe the data analysis idea corresponding to the second data analysis strategy. The data analysis idea can refer to a series of logical steps from clarifying the analysis goal to finally proposing decision-making suggestions during the data analysis process, which can include one or more of clarifying the analysis goal, data collection, data processing, selecting an analysis method, conducting data analysis, and summarization. The data analysis logic corresponding to the second data analysis strategy can be different from the data analysis logic corresponding to the preset strategy template. The data detail table corresponding to a certain data analysis logic can be understood as the detail table of the fields associated with this data analysis logic, and this data detail table can contain some or all of the fields associated with this data analysis logic.

[0087] Specifically, the knowledge item corresponding to the second data analysis strategy of the data analysis chapter can be obtained from the knowledge base of the data analysis chapter, such as obtaining the knowledge item related to the second data analysis strategy and the received data analysis requirements as the second knowledge item. According to the data analysis parameters, the second data analysis strategy, and the second knowledge item of the data analysis chapter, determine the data analysis logic corresponding to this second data analysis strategy. For example, input the data analysis goal of this data analysis chapter, the received data analysis requirements, and the second knowledge item into a preset model, and use the capabilities of the preset model to generate one or more other data analysis ideas different from the data analysis idea corresponding to the first data analysis strategy in the strategy template as the data analysis logic corresponding to the second data strategy.

[0088] For each generated data analysis logic, obtain the data table and field list associated with this data analysis logic from the meta-information library of this data analysis chapter, and generate a data detail table corresponding to this analysis logic based on this data table and field list, such as generating a data detail table containing at least some of the fields in this data table and field list. Among them, the data analysis chapter can be configured with a meta-information library, and this meta-information library can be used to store the meta-information of detailed data. For example, the data table and / or field structure information (such as field name, field type, and business meaning, etc.) related to this data analysis chapter can be stored in this meta-information library.

[0089] After generating the data detail table corresponding to a certain data analysis logic, data analysis can be performed based on this data detail table to generate a data report corresponding to the second data analysis strategy.

[0090] In the above embodiments, the method for performing data analysis based on the data detail list is not limited. For example, data can be extracted based on the data detail list and the extracted data can be analyzed based on this data detail list to obtain a data report, and so on.

[0091] In some embodiments, generating a data report corresponding to the second data analysis strategy based on the data detail list includes: respectively performing data extraction and summary calculation code generation based on the data detail list to obtain the extracted target data and the generated summary calculation code; using the summary calculation code to perform data summary calculation on the target data to obtain second report data; and performing data analysis on the second report data to obtain a second data analysis conclusion of the second report data.

[0092] Among them, the target data can be the data extracted based on the data detail list, such as the field values of each field extracted based on the detail list. The second report data can be understood as the report data generated based on the second data analysis strategy. The second data analysis conclusion can be understood as the analysis conclusion of the second report data.

[0093] Exemplarily, data can be extracted based on the data detail list, such as obtaining target data corresponding to the fields in the data detail list; and one or more data summary calculation codes can be generated based on the structure of the data detail list. The summary calculation result is obtained by performing summary calculation on the extracted target data using the generated summary calculation code and used as the second report data. Data analysis is performed on the generated second report data to obtain a second data analysis conclusion of the second report data.

[0094] S205. Stitch the data reports to obtain the chapter content of the data analysis chapter.

[0095] In this embodiment, after obtaining the data reports corresponding to one or more data analysis strategies of the data analysis chapter, the obtained data reports can be stitched to obtain the chapter content of this data analysis chapter.

[0096] S206. Stitch the chapter content of the data analysis chapter to obtain data analysis content that meets the data analysis requirements.

[0097] The data analysis method provided in this embodiment extracts the data analysis parameters of this data analysis chapter from the data analysis requirements and the data analysis content framework according to the parameter extraction strategy in the data analysis chapter, performs data analysis based on the extracted data analysis parameters and at least part of the data analysis strategies preset for this data analysis chapter, obtains the data reports corresponding to at least part of the data analysis strategies, and generates the chapter content of this data analysis chapter based on this data report, which can take into account both the generation efficiency and comprehensiveness of the generated chapter content, thereby improving the generation efficiency and comprehensiveness of the data analysis content.

[0098] Figure 3 It is a schematic diagram of the generation process of a data analysis content provided by an embodiment of the present disclosure. As Figure 3 shown, taking the data analysis content as a data report as an example, in some optional embodiments, the technical solution provided in this embodiment can be directed to specific data analysis scenarios and data structures, use a large language model (i.e., a preset model) and a data analysis calculation engine to accurately identify and understand the user's data analysis intention, quickly generate reusable data reports for vertical scenarios, and support modification or regeneration based on the existing data analysis content according to the modification opinions described in the user's natural language.

[0099] In this embodiment, the data report may include a report title (i.e., the title of the data content), a report analysis method and / or analysis idea (i.e., the content summary, used to describe the analysis method and related background of the overall data report), one or more data report chapters (including the problem breakdown of the data analysis target, the detailed data analysis and analysis conclusions of each sub-problem), and a data report conclusion (such as the summary of the overall data analysis report, including the answers, conclusions, and related suggestions to the user's original data analysis intention questions). Among them, each data report chapter may include a chapter title, a chapter description (description of the scenario and analysis method of the data analysis sub-problem, etc.), one or more data reports, and a chapter summary (such as the chapter summary obtained based on the report data content and chapter analysis theme of this chapter). Each data report may include a report title, report data (which may be data details, data charts, or multiple interrelated data indicators, etc.), and a report description (i.e., the data analysis conclusion, including the content introduction and brief analysis related to the data of this report)

[0100] In this embodiment, as Figure 3 shown, the generation of the data report mainly includes steps such as generating a data report framework (i.e., a data analysis content framework), generating chapter content, splicing the data report, and generating a data report conclusion. Among them, the generation of chapter content may include steps such as extracting chapter analysis parameters, extracting report data, analyzing report data, splicing chapters, and analyzing chapter data.

[0101] When generating a data report framework, exemplarily, data report intent recognition can be performed first. For example, according to the user's original description of data analysis (i.e., data analysis requirements), using the understanding and text extraction capabilities of a preset model, the user's key data analysis intents and target descriptions are extracted. Then, data analysis chapter selection is carried out. For example, according to the key data analysis intents and target descriptions extracted in the data report intent recognition stage, candidate data analysis chapters and corresponding strategies that match the user's data analysis description are screened from the chapter library, and after sorting according to the degree of relevance, several candidate data analysis chapters are selected as the data analysis chapters that make up the data report. After that, according to the selected data analysis chapters, the order of each data analysis chapter is arranged, and according to the data analysis themes of each data analysis chapter, the report title, report analysis methods, and / or analysis ideas of the data analysis report are generated. Then, the data report framework can be spliced. Among them, the chapter library can be pre-configured. The chapter library can contain the generation strategy configurations of all data analysis chapters that support generation. The generation strategy configuration of each data analysis chapter includes several templatized data analysis strategies (i.e., the first data analysis strategy), several personalized data analysis strategies (i.e., the second data analysis strategy), several parameter extraction strategies, and the relevant chapter knowledge base. A data analysis chapter can be an analysis of a specific data analysis scenario. For example, R & D quality analysis can be a chapter of the R & D data analysis content.

[0102] After the report framework generation is completed, the user is allowed to adjust or modify the generated data report framework. After the user confirms the report framework, the generation of each data analysis chapter in the data report framework to be generated will be triggered in parallel.

[0103] When generating chapter content, exemplarily, for the selected data analysis chapter, the generation strategy configuration of the corresponding chapter is loaded from the chapter library, and according to the user's data analysis requirements, the generation of the data analysis chapter content is completed in combination with the specific chapter generation strategy configuration.

[0104] Among them, the generation strategy configuration of each data analysis chapter can include the following system modules:

[0105] Scenario knowledge base: It contains knowledge entries related to data analysis in this scenario and the structure of the underlying data set. For example, the calculation formula of feature indicators, data grading strategies, relevant term concepts, data tables and field types in the data set, business meanings, etc.

[0106] Meta-information library: That is, the meta-information of the data system that actually stores detailed data, including data tables and field structure information (names, types, business meanings, etc.) related to the current data analysis chapter. The actual data calculation logic is generated based on the data asset meta-information.

[0107] Parameter extraction strategy: Based on this strategy, the large language model will extract the parameters required for the data analysis process from the user's original data analysis intention and the data report framework structure, such as the target time range, relevant personnel or project scope, and data statistical breakdown dimensions, etc.

[0108] Data report generation strategy template library: The template library contains common data report generation strategies for the target analysis scenarios of the data analysis chapter. For example, the common data analysis reports in the financial status analysis chapter can include the balance sheet, income statement, cash flow statement, etc. The detailed calculation and analysis strategies of these reports under specific data sets can be pre-configured in the template library as candidate data sets for the templatized data analysis generation part.

[0109] Personalized data report generation strategy: Completely based on the knowledge in the scenario knowledge base, without relying on the prefabricated report analysis strategies in the strategy template library, it generates personalized and open data analysis reports according to the user's intention. For example, when analyzing the financial status of a high-tech company, it can calculate and reasonably predict the future financial status based on the definitions of product R & D cycle and commercial customer acquisition ratio in the scenario knowledge base.

[0110] When generating the chapter content, exemplarily, the large language model can be used to extract the common parameters (i.e., data analysis parameters) required for the analysis chapter from the user's data analysis requirements as the input for the subsequent templatized data analysis steps and personalized data analysis steps of the strategy in this chapter. For example, if the user's original data analysis requirement is "How about the resource investment in the direction of user interaction optimization of XX business line in the third quarter of XXXX?", for the human resource investment analysis chapter, the extracted common parameters can be: target analysis time interval: from June 1, XXXX to September 30, XXXX, target analysis population: all personnel under XX business line; for the R & D quality analysis chapter, the extracted common parameters can be: target analysis time interval: from June 1, XXXX to September 30, XXXX, project scope to be statistically analyzed: projects related to user interaction optimization.

[0111] After the data analysis parameters are extracted, the templatized data analysis and personalized data analysis generation of the strategy configuration for this data analysis chapter can be triggered synchronously.

[0112] Among them, the process of templatized data analysis can be described as follows:

[0113] A1. Strategy template selection: Using the vectorization method or the capabilities of the large language model, screen out the data report generation templates in the strategy template library that are relevant to the user's data analysis intention and the chapter analysis target. The main screening process is as follows:

[0114] Rough screening of strategy templates: Use the embedding model to calculate the feature vectors of the user's data analysis intentions, and calculate the feature vectors of the data analysis objectives of this chapter in the data report framework; respectively use the above feature vectors to calculate the correlation with the analysis intentions and objectives of each data report generation template (i.e., strategy templates) in the strategy template library (i.e., the set of strategy templates), and obtain a set of candidate strategy templates;

[0115] Fine ranking of strategy templates: Utilize the capabilities of the large language model to judge the names, analysis intentions, and objectives of each data report in the set of candidate strategy templates, and select several data report generation strategies (i.e., preset strategy templates) that are most relevant to the user's data analysis intentions and the data analysis objectives of this chapter.

[0116] In this step, when there are few data report templates to be screened, fine ranking can be directly performed. Similarly, when the results of rough screening are few, fine ranking can be skipped.

[0117] A2. Rewriting of strategy templates: Utilize the understanding and rewriting capabilities of the large language model (i.e., the preset model) to simply rewrite the selected strategy templates. The specific rewriting process is described as follows:

[0118] Select relevant knowledge entries (i.e., the first knowledge entries), relevant data reports, and fields from the knowledge base that are related to the preset strategy template and the user's analysis intentions; read the data indicators, dimension calculation formulas, relevant data report fields, and data calculation codes related to the data report from the preset strategy template; utilize the large language model, refer to the relevant knowledge entries, and rewrite the relevant data indicator and dimension calculation formula codes of the preset strategy template. For example, if the user needs to analyze the financial status of each subsidiary in Region A, then the calculation and analysis strategies of prefabricated statements such as the balance sheet, income statement, and cash flow statement can be adjusted to generate new data calculation codes and analysis strategies.

[0119] A3. Data calculation and analysis: Based on the rewritten strategy template, use the stored data for summary calculation and complete the corresponding data result analysis.

[0120] The process of personalized data analysis can be described as follows:

[0121] B1. Extract relevant knowledge entries from the knowledge base that are related to the strategy template and the user's analysis intentions.

[0122] B2. Based on the data analysis objectives of this chapter and the user's original analysis intentions (i.e., data analysis requirements), utilize the capabilities of the large language model to generate other data analysis ideas (i.e., data analysis logic) in addition to templatized data analysis.

[0123] B3. Select data tables and field lists related to the data analysis idea from the meta-information database, and generate a data details table, which can include all fields related to the data analysis idea.

[0124] B4. Pull detailed data based on the data details table; and, based on the data details table, generate several data analysis calculation logics (such as summary calculation codes).

[0125] B5. Perform summary calculations based on the dynamically generated data analysis calculation logics.

[0126] B6. Analyze the data analysis results to obtain partial conclusions of personalized data analysis.

[0127] After the content generation of each data chapter is completed, the content of the data report can be spliced according to the design of the data inclusion framework, and the overall conclusion of the data report can be summarized and generated based on the conclusions of each data analysis chapter.

[0128] This embodiment supports pre-configuring the calculation strategies and related knowledge for actual data analysis scenarios, using a large language model to disassemble the actual data analysis target, and fine-tuning and rewriting predefined data analysis strategies with high accuracy according to the user's actual data analysis intention, greatly improving the accuracy of data report generation; in addition, it also makes full use of the large language model's ability to understand implicit information in special scenarios, and uses a personalized data analysis process to generate a data analysis idea strongly related to the user's actual problem; finally, each data analysis chapter in the overall data report combines the above two parts of data analysis conclusions to form a complete data report. The data analysis report generated by using the solution of this embodiment has a relatively high and controllable data accuracy rate, and the semantic understanding and code writing capabilities of the large language model are significantly lower than those of the technical solution that completely relies on the large language model to generate a complete data report. The overall data report generation process supports multiple rounds of interaction modification and fine-tuning with the user, significantly reducing the engineering difficulty of the large language model data report generation process.

[0129] Figure 4 It is a structural block diagram of a data analysis device provided by an embodiment of the present disclosure. This device can be implemented by software and / or hardware, and can be configured in an electronic device. Typically, it can be configured in a computer, a mobile phone or a tablet computer, and can generate data analysis content based on the data analysis requirements input by the user by executing a data analysis method. As Figure 4 shown, the data analysis device provided by this embodiment may include: an information acquisition module 401, a framework generation module 402, a chapter generation module 403, and a chapter splicing module 404, where

[0130] An information acquisition module 401 is configured to select a data analysis chapter that matches the received data analysis requirement from a chapter library, and acquire chapter information of the data analysis chapter, where the chapter information includes chapter analysis attribute information and a chapter generation strategy set;

[0131] A framework generation module 402 is configured to generate a data analysis content framework based on the chapter analysis attribute information and the chapter generation strategy set of the data analysis chapter;

[0132] A chapter generation module 403 is configured to generate chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation strategy set of the data analysis chapter, and the data analysis requirement;

[0133] A chapter splicing module 404 is configured to splice the chapter content of the data analysis chapter to obtain data analysis content that meets the data analysis requirement.

[0134] The data analysis apparatus provided in this embodiment selects a data analysis chapter that matches the received data analysis requirement from a chapter library through the information acquisition module, and acquires chapter information of the selected data analysis chapter, where the chapter information includes chapter analysis attribute information and a chapter generation strategy set; generates a data analysis content framework based on the chapter analysis attribute information and the chapter generation strategy set of the data analysis chapter through the framework generation module; generates chapter content of each data analysis chapter according to the data analysis content framework, the chapter generation strategy of the data analysis chapter, and the received data analysis requirement through the chapter generation module; and splices the chapter content of each data analysis chapter through the chapter splicing module to obtain data analysis content that meets the received data analysis requirement. By using the above technical solution, this embodiment pre-configures a chapter library and configures a chapter generation strategy for each chapter supported in the chapter library, and after receiving a data analysis requirement, selects a data analysis chapter that matches the received data analysis requirement from the chapter library to generate a data analysis content framework, and generates chapter content of each data analysis chapter in the data analysis content framework according to the chapter generation strategy of each data analysis chapter, and can automatically generate data analysis content that meets the data analysis requirement based on the pre-configured strategy. Compared with the technical solutions of manually performing data mining and analysis or completely relying on a large language model to generate data analysis content, the generation efficiency and accuracy of the data analysis content can be improved.

[0135] Optionally, the chapter generation strategy set includes a parameter extraction strategy subset and a data analysis strategy subset, where the parameter extraction strategy subset includes at least one parameter extraction strategy; the data analysis strategy subset includes a first data analysis strategy based on a strategy template and / or a second data analysis strategy not based on a strategy template.

[0136] Optionally, the chapter generation module 403 may include: a parameter extraction unit configured to extract data analysis parameters of the data analysis chapter from the data analysis requirements and the data analysis content framework according to the parameter extraction strategy of the data analysis chapter; a data analysis unit configured to perform data analysis based on the data analysis parameters and at least part of the data analysis strategies of the data analysis chapter to obtain a data report corresponding to the at least part of the data analysis strategies, where the data report includes report data and a data analysis conclusion; and a report splicing unit configured to splice the data reports to obtain the chapter content of the data analysis chapter.

[0137] Optionally, the data analysis unit may include: a template selection subunit configured to select a preset strategy template matching the data analysis requirements from a set of strategy templates of the data analysis chapter, where the preset strategy template includes a pre-set first data analysis strategy and a data extraction code corresponding to the first data analysis strategy; a template adjustment subunit configured to adjust the preset strategy template based on the data analysis parameters and a first knowledge item, where the first knowledge item is a knowledge item corresponding to the preset strategy template in the knowledge base of the data analysis chapter; and a data analysis subunit configured to perform data analysis using the adjusted preset strategy template to obtain a data report corresponding to the preset strategy template.

[0138] Optionally, the template selection subunit may be specifically configured to: calculate the matching degree between the strategy templates in the set of strategy templates and the data analysis requirements by using at least one matching degree calculation method, and select at least one strategy template from the set of strategy templates as the preset strategy template matching the data analysis requirements, where the at least one matching method includes a vector matching method and / or a semantic matching method.

[0139] Optionally, the data analysis subunit may be specifically configured to: perform data extraction by using the adjusted data extraction code in the preset strategy template to obtain first report data corresponding to the preset strategy template; and perform data analysis on the first report data based on the adjusted first data analysis strategy in the preset strategy template to obtain a first data analysis conclusion of the first report data.

[0140] Optionally, the data analysis unit may include: a logic determination subunit, configured to determine the data analysis logic of the data analysis chapter according to the data analysis parameter, the second data analysis strategy, and the second knowledge item, where the second knowledge item is the knowledge item corresponding to the second data analysis logic in the knowledge base of the data analysis chapter; a detailed list generation subunit, configured to generate a data detailed list corresponding to the data analysis logic, where the data detailed list at least indicates the associated fields of the data analysis logic; and a data report generation subunit, configured to generate a data report corresponding to the second data analysis strategy based on the data detailed list.

[0141] Optionally, the data report generation subunit may be specifically configured to: respectively perform data extraction and summary calculation code generation based on the data detailed list, obtain the extracted target data and the generated summary calculation code; use the summary calculation code to perform data summary calculation on the target data to obtain second report data; and perform data analysis on the second report data to obtain a second data analysis conclusion of the second report data.

[0142] Optionally, the framework generation module 402 may be specifically configured to: generate the chapter framework information of the data analysis chapter according to the chapter analysis attribute information and the chapter generation strategy set of the data analysis chapter; sort the chapter framework information according to the matching degree between the data analysis chapter and the data analysis requirements to obtain a sorting result of the chapter framework information; generate a content title and a content abstract of the data analysis content according to the sorting result and the chapter analysis attribute information of the data analysis chapter; and splice the content title, the content abstract, and the chapter framework information of the data analysis chapter to obtain the content framework of the data analysis content.

[0143] The data analysis device provided in the embodiments of the present disclosure can execute the data analysis method provided in any embodiment of the present disclosure, and has function modules and beneficial effects corresponding to executing the data analysis method. For technical details not described in detail in this embodiment, reference may be made to the data analysis method provided in any embodiment of the present disclosure.

[0144] Reference is made below to Figure 5 FIG. which shows a schematic structural diagram of an electronic device (such as a terminal device or a server) 500 suitable for implementing the embodiments of the present disclosure. The terminal device in the embodiments of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5The electronic device shown is merely an example and should not impose any limitation on the functions and scope of use of the embodiments of the present disclosure.

[0145] As Figure 5 shown, the electronic device 500 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 501, which may perform various appropriate actions and processes according to the programs stored in the read-only memory (ROM) 502 or the programs loaded from the storage device 508 into the random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. The input / output (I / O) interface 505 is also connected to the bus 504.

[0146] Generally, the following devices may be connected to the I / O interface 505: an input device 506 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 508 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 509. The communication device 509 may allow the electronic device 500 to communicate with other devices wirelessly or wiredly to exchange data. Although Figure 5 the electronic device 500 with various devices is shown, it should be understood that it is not required to implement or have all the shown devices. More or fewer devices may be implemented or had alternatively.

[0147] In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, the embodiments of the present disclosure include a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes program codes for performing the methods shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from the network through the communication device 509, or installed from the storage device 508, or installed from the ROM 502. When the computer program is executed by the processing device 501, the above-mentioned functions defined in the methods of the embodiments of the present disclosure are executed.

[0148] It should be noted that the above-mentioned computer-readable medium in the present disclosure can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present disclosure, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and this computer-readable signal medium can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0149] In some embodiments, the client and the server can communicate using any currently known or future-developed network protocol such as HTTP (HyperText Transfer Protocol), and can be interconnected with digital data communication in any form or medium (for example, a communication network). Examples of communication networks include local area networks ("LAN"), wide area networks ("WAN"), the Internet (for example, the Internet), and end-to-end networks (for example, ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0150] The above-mentioned computer-readable medium can be included in the above-mentioned electronic device; it can also exist separately and not be assembled into the electronic device.

[0151] The above computer-readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: select, from a chapter library, a data analysis chapter that matches the received data analysis requirement, and obtain chapter information of the data analysis chapter, where the chapter information includes chapter analysis attribute information and a chapter generation strategy set; generate a data analysis content framework based on the chapter analysis attribute information and the chapter generation strategy set of the data analysis chapter; generate the chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation strategy set of the data analysis chapter, and the data analysis requirement; splice the chapter content of the data analysis chapter to obtain data analysis content that meets the data analysis requirement.

[0152] Computer program code for carrying out operations of the present disclosure may be written in one or more programming languages or combinations thereof. The programming languages include, but are not limited to, object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0153] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a portion of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that, in some alternative implementations, the functions denoted by the blocks may occur in an order different from that denoted in the drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0154] The units involved in the embodiments of the present disclosure can be implemented in software or in hardware. Among them, the name of the module does not constitute a limitation to the unit itself in some cases.

[0155] The functions described above in this article can be performed at least in part by one or more hardware logic components. For example, without limitation, the exemplary types of hardware logic components that can be used include: Field Programmable Gate Array (FPGA), Application Specific Integrated Circuit (ASIC), Application Specific Standard Product (ASSP), System on Chip (SOC), Complex Programmable Logic Device (CPLD), and so on.

[0156] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include electrical connections based on one or more wires, portable computer disks, hard disks, Random Access Memory (RAM), Read Only Memory (ROM), Erasable Programmable Read Only Memory (EPROM or Flash Memory), optical fibers, portable compact disc read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0157] According to one or more embodiments of the present disclosure, Example 1 provides a data analysis method, including:

[0158] Select a data analysis chapter that matches the received data analysis requirement from the chapter library, and obtain the chapter information of the data analysis chapter, where the chapter information includes chapter analysis attribute information and a chapter generation policy set;

[0159] Generate a data analysis content framework based on the chapter analysis attribute information and the chapter generation policy set of the data analysis chapter;

[0160] Generate the chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation policy set of the data analysis chapter, and the data analysis requirement;

[0161] Stitch together the chapter content of the data analysis chapter to obtain data analysis content that meets the data analysis requirement.

[0162] According to one or more embodiments of the present disclosure, Example 2 Based on the method described in Example 1, the chapter generation policy set includes a parameter extraction policy subset and a data analysis policy subset, wherein the parameter extraction policy subset includes at least one parameter extraction policy; the data analysis policy subset includes a first data analysis policy based on a policy template and / or a second data analysis policy not based on a policy template.

[0163] According to one or more embodiments of the present disclosure, Example 3 Based on the method described in Example 2, generating the chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation policy set of the data analysis chapter, and the data analysis requirements includes:

[0164] Extract the data analysis parameters of the data analysis chapter from the data analysis requirements and the data analysis content framework according to the parameter extraction policy of the data analysis chapter;

[0165] Perform data analysis based on the data analysis parameters and at least part of the data analysis policies of the data analysis chapter to obtain a data report corresponding to the at least part of the data analysis policies, where the data report includes report data and a data analysis conclusion;

[0166] Stitch the data reports to obtain the chapter content of the data analysis chapter.

[0167] According to one or more embodiments of the present disclosure, Example 4 Based on the method described in Example 3, performing data analysis based on the data analysis parameters and at least part of the data analysis policies of the data analysis chapter to obtain a data report corresponding to the at least part of the data analysis policies includes:

[0168] Select a preset policy template that matches the data analysis requirements from the policy template set of the data analysis chapter, where the preset policy template includes a pre-set first data analysis policy and the data extraction code corresponding to the first data analysis policy;

[0169] Adjust the preset policy template based on the data analysis parameters and the first knowledge item, where the first knowledge item is the knowledge item corresponding to the preset policy template in the knowledge base of the data analysis chapter;

[0170] Perform data analysis using the adjusted preset policy template to obtain a data report corresponding to the preset policy template.

[0171] According to one or more embodiments of the present disclosure, Example 5 Based on the method described in Example 4, selecting a preset policy template that matches the data analysis requirements from the policy template set of the data analysis chapter includes:

[0172] Calculate the matching degree between the policy templates in the policy template set and the data analysis requirements by using at least one matching degree calculation method, and select at least one policy template from the policy template set according to the matching degree as the preset policy template that matches the data analysis requirements, where the at least one matching method includes a vector matching method and / or a semantic matching method.

[0173] According to one or more embodiments of the present disclosure, Example 6 is based on the method described in Example 4. The data analysis is performed using the adjusted preset policy template to obtain a data report corresponding to the preset policy template, including:

[0174] Extract data using the adjusted data extraction code in the preset policy template to obtain first report data corresponding to the preset policy template;

[0175] Perform data analysis on the first report data based on the adjusted first data analysis strategy in the preset policy template to obtain a first data analysis conclusion of the first report data.

[0176] According to one or more embodiments of the present disclosure, Example 7 is based on the method described in Example 3. The data analysis is performed based on at least part of the data analysis strategies of the data analysis parameters and the data analysis chapter to obtain a data report corresponding to the at least part of the data analysis strategies, including:

[0177] Determine the data analysis logic of the data analysis chapter according to the data analysis parameters, the second data analysis strategy, and the second knowledge item, where the second knowledge item is the knowledge item corresponding to the second data analysis logic in the knowledge base of the data analysis chapter;

[0178] Generate a data detail table corresponding to the data analysis logic, where the data detail table at least indicates the associated fields of the data analysis logic;

[0179] Generate a data report corresponding to the second data analysis strategy based on the data detail table.

[0180] According to one or more embodiments of the present disclosure, Example 8 is based on the method described in Example 7. The data report corresponding to the second data analysis strategy is generated based on the data detail table, including:

[0181] Respectively perform data extraction and summary calculation code generation based on the data detail table to obtain the extracted target data and the generated summary calculation code;

[0182] Perform data summary calculation on the target data using the summary calculation code to obtain second report data;

[0183] Perform data analysis on the second report data to obtain a second data analysis conclusion of the second report data.

[0184] According to one or more embodiments of the present disclosure, Example 9 According to the method described in any one of Examples 1-8, the method for generating a data analysis content framework based on the chapter analysis attribute information and chapter generation policy set of the data analysis chapter includes:

[0185] Generate chapter framework information for the data analysis chapter according to the chapter analysis attribute information and chapter generation policy set of the data analysis chapter;

[0186] Sort the chapter framework information according to the matching degree between the data analysis chapter and the data analysis requirements to obtain a sorting result of the chapter framework information;

[0187] Generate a content title and a content abstract for the data analysis content based on the sorting result and the chapter analysis attribute information of the data analysis chapter;

[0188] Concatenate the content title, the content abstract, and the chapter framework information of the data analysis chapter to obtain the content framework of the data analysis content.

[0189] According to one or more embodiments of the present disclosure, Example 10 provides a data analysis device, including:

[0190] An information acquisition module, configured to select a data analysis chapter that matches the received data analysis requirements from a chapter library, and acquire chapter information of the data analysis chapter, where the chapter information includes chapter analysis attribute information and a chapter generation policy set;

[0191] A framework generation module, configured to generate a data analysis content framework based on the chapter analysis attribute information and chapter generation policy set of the data analysis chapter;

[0192] A chapter generation module, configured to generate chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation policy set of the data analysis chapter, and the data analysis requirements;

[0193] A chapter concatenation module, configured to concatenate the chapter content of the data analysis chapter to obtain data analysis content that meets the data analysis requirements.

[0194] According to one or more embodiments of the present disclosure, Example 11 provides an electronic device, including:

[0195] One or more processors;

[0196] A memory, configured to store one or more programs,

[0197] When the one or more programs are executed by the one or more processors, the one or more processors implement the data analysis method as described in any one of Examples 1-9.

[0198] According to one or more embodiments of the present disclosure, Example 12 provides a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, it implements the data analysis method as described in any one of Examples 1-9.

[0199] According to one or more embodiments of the present disclosure, Example 13 provides a computer program product, and when the computer program product is executed by a computer, the computer implements the data analysis method as described in any one of Examples 1-9.

[0200] The above description is only a preferred embodiment of the present disclosure and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the present disclosure.

[0201] In addition, although the operations are depicted in a particular order, this should not be construed as requiring that the operations be performed in the particular order shown or in sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present disclosure. Certain features described in the context of separate embodiments may also be implemented combinatorially in a single embodiment. Conversely, the various features described in the context of a single embodiment may also be implemented separately or in any suitable sub-combination in multiple embodiments.

[0202] Although the subject matter has been described in language specific to structural features and / or methodological logical acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely example forms for implementing the claims.

Claims

1. A data analysis method, characterized in that: include: Selecting a data analysis chapter matching the received data analysis requirement from a chapter library, and acquiring chapter information of the data analysis chapter, wherein the chapter information includes chapter analysis attribute information and a chapter generation strategy set; Generate a data analysis content framework based on the chapter analysis attribute information and chapter generation strategy set of the data analysis chapter; Generate chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation strategy set of the data analysis chapter, and the data analysis requirements; The chapter contents of the data analysis chapter are spliced ​​to obtain data analysis content that meets the data analysis requirements.

2. The method according to claim 1, characterized in that The chapter generation strategy set includes a parameter extraction strategy subset and a data analysis strategy subset, wherein the parameter extraction strategy subset includes at least one parameter extraction strategy; the data analysis strategy subset includes a first data analysis strategy based on a strategy template and / or a second data analysis strategy not based on a strategy template.

3. The method according to claim 2, characterized in that Generating the chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation strategy set of the data analysis chapter and the data analysis requirements includes: Extracting data analysis parameters of the data analysis chapter from the data analysis requirements and the data analysis content framework according to the parameter extraction strategy of the data analysis chapter; Perform data analysis based on the data analysis parameters and at least part of the data analysis strategy of the data analysis section to obtain a data report corresponding to at least part of the data analysis strategy, wherein the data report includes report data and data analysis conclusions; The data reports are spliced ​​to obtain the chapter content of the data analysis chapter.

4. The method according to claim 3, characterized in that The performing of data analysis based on the data analysis parameters and at least part of the data analysis strategy of the data analysis section to obtain a data report corresponding to at least part of the data analysis strategy includes: Selecting a preset strategy template that matches the data analysis requirement from the strategy template set in the data analysis chapter, wherein the preset strategy template includes a preset first data analysis strategy and a data extraction code corresponding to the first data analysis strategy; Adjusting the preset strategy template based on the data analysis parameter and a first knowledge item, wherein the first knowledge item is a knowledge item in the knowledge base of the data analysis section corresponding to the preset strategy template; The adjusted preset strategy template is used to perform data analysis to obtain a data report corresponding to the preset strategy template.

5. The method according to claim 4, characterized in that The selecting a preset policy template matching the data analysis requirement from the policy template set in the data analysis section includes: At least one matching degree calculation method is used to calculate the matching degree between the policy templates in the policy template set and the data analysis requirements, and at least one policy template is selected from the policy template set according to the matching degree as a preset policy template that matches the data analysis requirements, wherein the at least one matching method includes a vector matching method and / or a semantic matching method.

6. The method according to claim 4, characterized in that The data analysis is performed using the adjusted preset strategy template to obtain a data report corresponding to the preset strategy template, including: Using the adjusted data extraction code in the preset policy template to extract data, and obtaining first report data corresponding to the preset policy template; The first report data is analyzed based on the adjusted first data analysis strategy in the preset strategy template to obtain a first data analysis conclusion of the first report data.

7. The method according to claim 3, characterized in that The performing of data analysis based on the data analysis parameters and at least part of the data analysis strategy of the data analysis section to obtain a data report corresponding to at least part of the data analysis strategy includes: Determine the data analysis logic of the data analysis chapter according to the data analysis parameter, the second data analysis strategy, and a second knowledge item, wherein the second knowledge item is a knowledge item in the knowledge base of the data analysis chapter corresponding to the second data analysis logic; Generate a data detail table corresponding to the data analysis logic, wherein the data detail table at least indicates associated fields of the data analysis logic; A data report corresponding to the second data analysis strategy is generated based on the data detail table.

8. The method according to claim 7, characterized in that The step of generating a data report corresponding to the second data analysis strategy based on the data detail table includes: Extracting data and generating summary calculation codes based on the data detail table respectively, and obtaining the extracted target data and the generated summary calculation codes; Using the summary calculation code to perform data summary calculation on the target data to obtain second report data; Perform data analysis on the second report data to obtain a second data analysis conclusion of the second report data.

9. The method according to any one of claims 1 to 8, characterized in that: The step of generating a data analysis content framework based on the chapter analysis attribute information and the chapter generation strategy set of the data analysis chapter includes: Generating chapter framework information of the data analysis chapter according to the chapter analysis attribute information and the chapter generation strategy set of the data analysis chapter; Sorting the chapter framework information according to the matching degree between the data analysis chapter and the data analysis requirement to obtain a sorting result of the chapter framework information; Generating a content title and a content summary of the data analysis content according to the sorting result and the chapter analysis attribute information of the data analysis chapter; The content title, the content summary and the chapter framework information of the data analysis chapter are spliced ​​to obtain the content framework of the data analysis content.

10. A data analysis device, characterized in that: include: An information acquisition module, used for selecting a data analysis chapter matching the received data analysis requirement from the chapter library, and acquiring chapter information of the data analysis chapter, wherein the chapter information includes chapter analysis attribute information and a chapter generation strategy set; A framework generation module, used to generate a data analysis content framework based on the chapter analysis attribute information and chapter generation strategy set of the data analysis chapter; A chapter generation module, used for generating chapter content of the data analysis chapter according to the data analysis content framework, the chapter generation strategy set of the data analysis chapter and the data analysis requirements; The chapter splicing module is used to splice the chapter contents of the data analysis chapters to obtain data analysis contents that meet the data analysis requirements.

11. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the data analysis method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the data analysis method according to any one of claims 1 to 9 when executed.

13. A computer program product, characterized in that The computer program product comprises a computer program, which, when executed by a processor, implements the data analysis method according to any one of claims 1 to 9.

Images