Cross-theme data analysis method and device, medium and product

Abstract

The embodiment of the invention provides a cross-theme data analysis method and device, a medium and a product, and aims to improve the accuracy of an analysis result for an analysis target. The method comprises the following steps: collecting data of different themes in one or more fields associated with an analysis target; according to a preset identifier, the data of the different themes are split into a plurality of data sets, the identifier is independent of any single theme, and each data set comprises partial data of the multiple themes associated with the same value of the identifier in the data of the different themes; and obtaining an analysis result aiming at the analysis target by performing independent analysis on the interaction relationship among the data of different themes on the plurality of data sets.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a cross-topic data analysis method, device, medium and product. Background Art

[0002] In data analysis and mining technology, different topics are usually designed according to specific research questions or application scenarios. By classifying and organizing the collected data by topic, the complexity of the data can be reduced, thereby improving the efficiency of data analysis.

[0003] However, due to the high coupling of data under various topics, traditional data analysis methods usually obtain analysis results based on data of a single topic, resulting in low accuracy of analysis results. Taking market trend prediction as an example, traditional data analysis methods usually perform analysis based on data of a single topic such as product popularity evaluation, and then use the analysis results obtained from the data of the single topic as the analysis results for the market trend prediction. Summary of the invention

[0004] In view of the above problems, the embodiments of the present application provide a cross-topic data analysis method, device, medium and product to overcome the above problems or at least partially solve the above problems.

[0005] In a first aspect of an embodiment of the present application, a cross-topic data analysis method is provided, the method comprising: Collect data on different topics under one or more areas related to the analysis objectives; According to a preset identifier, the data of the different subjects are divided into multiple data sets, wherein the identifier is independent of any single subject, and each of the data sets includes: in the data of the different subjects, partial data of multiple subjects associated with the same value of the identifier; By independently analyzing the interaction relationship between data of different subjects on the multiple data sets, analysis results targeting the analysis target are obtained.

[0006] As a possible implementation, the analysis result for the analysis target is obtained by independently analyzing the interaction relationship between data of different subjects on the multiple data sets, including: Grouping the multiple data sets to obtain multiple data groups; By independently analyzing the interaction relationship between data of different topics on the multiple data groups, analysis results targeting the analysis target are obtained.

[0007] As a possible implementation, the analysis result for the analysis target is obtained by independently analyzing the interaction relationship between data of different subjects on the multiple data groups, including: For each of the data groups, performing independent analysis on the interaction relationship between data of different subjects on the data group; For each of the data groups, according to the interaction relationship between the data of different topics analyzed, data analysis on different topics is performed on the data groups respectively; The data analysis results obtained for each of the plurality of data groups are determined as analysis results for the analysis target.

[0008] As a possible implementation mode, according to the interaction relationship between the data of different themes analyzed, the data groups are respectively subjected to data analysis for different themes, including: Extracting characteristic data corresponding to each of the different topics from the data group; According to the analyzed interaction relationship between the data of different topics, the preset analysis methods corresponding to the different topics are used to perform data analysis on the characteristic data corresponding to the different topics.

[0009] As a possible implementation manner, grouping the multiple data sets to obtain multiple data groups includes: The plurality of data sets are grouped by a clustering algorithm according to the degree of feature similarity between objects of different categories associated with the data sets to obtain the plurality of data groups.

[0010] As a possible implementation, the features of an object associated with a single data set are described by a feature vector, and the feature vector is determined by the following steps: Preprocessing the data set, wherein the preprocessing includes at least one of data cleaning and standardization; From the preprocessed data set, multidimensional features of the objects associated with the data set are extracted to obtain the feature vector.

[0011] As a possible implementation manner, before splitting the data of different topics into multiple data sets according to the preset identifier, the method further includes: Extracting target objects from the data of different subjects, wherein the target objects include: the same or related objects associated with the data of multiple subjects; Identifiers with different values ​​are generated for different categories of the target object, and the identifiers with different values ​​are stored in a database.

[0012] According to a second aspect of the embodiments of the present application, a cross-topic data analysis device is provided, the device comprising: A data collection module is used to collect data on different topics under one or more fields related to the analysis target; A data decoupling module, used for splitting the data of different subjects into multiple data sets according to a preset identifier, wherein the identifier is independent of any single subject, and each of the data sets includes: in the data of different subjects, partial data of multiple subjects associated with the same value of the identifier; The data analysis module is used to obtain analysis results for the analysis target by independently analyzing the interaction relationship between data of different subjects on the multiple data sets.

[0013] According to a third aspect of an embodiment of the present application, there is provided an electronic device comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; the memory is used to store computer programs; and the processor is used to implement the cross-subject data analysis method as described in the first aspect of the embodiment of the present application when executing the program stored in the memory.

[0014] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the cross-subject data analysis method as described in the first aspect disclosed in the embodiment of the present application is implemented.

[0015] According to a fifth aspect of an embodiment of the present application, a computer program product is provided, including a computer program / instruction, which, when executed by a processor, implements the cross-subject data analysis method as described in the first aspect of the embodiment of the present application.

[0016] The embodiments of the present application include the following advantages: using identifiers independent of any single subject to split and associate data of multiple subjects, thereby decoupling and reorganizing the data within each subject based on the correlation between the various parts of the data within different subjects, thereby providing rich data resources of multiple subjects for data analysis through the obtained data set; then, by performing independent cross-subject analysis on multiple data sets, interference between data sets can be avoided, thereby achieving a more refined cross-subject data analysis process, which can greatly improve the accuracy of the final analysis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0018] Figure 1 This is a flowchart of the steps of a cross-topic data analysis method in an embodiment of the present application; Figure 2 It is a schematic diagram of an implementation process of a data analysis method in an embodiment of the present application; Figure 3 It is a structural schematic diagram of a cross-topic data analysis device according to an embodiment of the present application; Figure 4 It is a schematic diagram of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0019] First of all, in order to facilitate understanding of the technical solutions provided by the present application, the main technical concepts involved in the embodiments of the present application are briefly described below.

[0020] Field: refers to a broad range of research or applications such as social science research, market analysis, public health monitoring, etc. Generally speaking, a field will have multiple topics. Different topics in a field may share some basic data or analysis methods, but their problems and focuses are different.

[0021] Topic: refers to specific research problems or application scenarios that are further subdivided within a field. For example, in the field of market analysis, there may be topics such as consumer behavior analysis and product popularity evaluation.

[0022] Population package: data of a specific user group obtained by segmenting the population data according to the specific characteristics of the users.

[0023] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0024] Reference Figure 1 As shown, a flowchart of a cross-topic data analysis method in an embodiment of the present application is shown, and the method includes the following steps: Step S11: Collect data on different topics in one or more fields associated with the analysis target.

[0025] Among them, the analysis objectives can be user behavior pattern recognition, health risk assessment, market trend prediction, etc., which can be flexibly set by the user according to actual data analysis needs.

[0026] In specific implementation, the fields and topics associated with the analysis target can be determined based on experience. For example, when market trend forecasting is used as the analysis target, topics such as consumer behavior analysis and product popularity evaluation in the market analysis field can be used as different topics associated with the analysis target, and data on these topics can be collected.

[0027] Alternatively, the relevance between the analysis target and the data of different topics in various fields can also be determined based on whether the objects associated with the data of different topics in various fields are the same or related to the objects associated with the analysis target. For example, when the analysis target is user behavior pattern recognition, the object associated with it is the user, then the data of different topics in various fields associated with the user, that is, the data of different topics in various fields containing user-related information, such as shopping records, social media interaction records, health data, etc., can be used as the data of different topics associated with the analysis target.

[0028] Step S12: split the data of the different topics into multiple data sets according to a preset identifier, wherein the identifier is independent of any single topic, and each data set includes: partial data of multiple topics associated with the same value of the identifier in the data of the different topics.

[0029] In specific implementation, an identifier independent of any single subject is pre-set, that is, to avoid setting an identifier based on the unique data of a certain subject, thereby ensuring that the preset identifier can associate data of multiple subjects.

[0030] Taking the identifier as a crowd package identifier as an example, the different values ​​0 and 1 of the crowd package identifier are used to identify different user groups A1 and B1 respectively. Then, according to the crowd package identifier, the crowd data of different topics can be split into crowd package A2 associated with crowd package identifier 0 and crowd package B2 associated with crowd package identifier 1 (i.e., split into two data sets), thereby reducing the coupling degree of data of different topics. The crowd package A2 includes: part of the data associated with user group A1 in the crowd data of each topic, and the crowd package B2 includes: part of the data associated with user group B1 in the crowd data of each topic.

[0031] Step S13: Obtain analysis results for the analysis target by independently analyzing the interaction relationship between data of different subjects on the multiple data sets.

[0032] In specific implementation, each data set is independently analyzed for the interaction relationship between data of different topics (such as causal relationship, positive correlation and negative correlation, etc.). For example, each data set can be input into a pre-trained machine learning model, and the interaction relationship between data of different topics can be captured by the machine learning model. Then, the data of different topics within the data set are comprehensively analyzed according to the interaction relationship. For example, after setting data weights for data with causal relationships in each data set or verifying their reliability, comprehensive data analysis is performed on each data set for the analysis target to obtain the final analysis result. In this way, with the help of the richness of cross-topic data resources and the interaction relationship between data, a more in-depth and accurate data analysis of the analysis target can be achieved.

[0033] Optionally, after obtaining the analysis results for the analysis target, the analysis results may be visualized, or relevant business strategies may be adjusted and optimized more scientifically and effectively based on the analysis results.

[0034] It is understandable that the use of independent analysis can avoid interference between different data sets (such as population packages of different user groups), and thus the scope of analysis can be finely controlled based on the data set to be within partial data of different topics that have some association (such as association with the same user group). This can more accurately capture the interaction between data of different topics, thereby improving the accuracy of the final analysis results.

[0035] By adopting the technical solution of the embodiment of the present application, the data of multiple topics are split and associated using identifiers independent of any single topic, thereby decoupling and reorganizing the data within each topic based on the correlation between the various parts of the data within different topics, thereby providing rich data resources of multiple topics for data analysis through the obtained data set; then, by performing independent cross-topic analysis of multiple data sets, interference between data sets can be avoided, thereby achieving a more refined cross-topic data analysis process, which can greatly improve the accuracy of the final analysis results.

[0036] As a possible implementation, in the step S12, before splitting the data of different topics into multiple data sets according to a preset identifier, the method further includes: Step S14: extracting target objects from the data of different subjects, wherein the target objects include: identical or related objects associated with data of multiple subjects.

[0037] In specific implementation, refer to Figure 2The schematic diagram of the implementation process of the data analysis method shown in FIG. 1 shows that after the information system collects data of different topics, it can be input into the data analysis system through the application programming interface (API) or the file system. The data analysis system can then automatically identify the objects and their classifications associated with the data of different topics by using methods such as keyword detection.

[0038] For example, for the data of each topic, it is possible to determine whether the data is associated with objects such as users, user groups, devices or events by detecting whether there are keywords such as device identification, event number or user account in the data, and determine the classification of the associated object based on the value of the detected keyword.

[0039] After determining the objects and their classifications associated with the data of each topic, find the same or related objects as the target objects. For example, if the data of topics A and B are associated with different types of devices (corresponding to the case of associating the same object), the device can be used as the target object, and different identifiers will be generated for different types of devices later; for another example, if the data of topic A is associated with users, and the data of topic B is associated with user groups (corresponding to the case of associating related objects), users or user groups can be used as target objects according to actual needs, and different identifiers will be generated for different users or user groups later.

[0040] Step S15: generating identifiers with different values ​​for different categories of the target object, and storing the identifiers with different values ​​in a database.

[0041] In specific implementation, for each target object of each category, a universally unique identifier (UUID) generated based on data such as the current time and hardware identification can be used as the identifier value corresponding to the target object of the category, thereby ensuring that identifiers with different values ​​can uniquely identify the target object of the corresponding category. The generated identifier is then stored in a database so that the identifier can be quickly queried and used in subsequent data analysis steps.

[0042] In this embodiment, considering that data analysis is usually performed on the properties of objects of specific categories (such as specific users, specific devices or specific events, etc.), the embodiment of the present application generates identifiers based on the same or related objects and their categories associated with data of multiple topics, and uses the identifiers to split and associate data of different topics, which facilitates subsequent data analysis around objects of different categories, thereby better meeting actual data analysis needs.

[0043] As a possible implementation, in step S13, by independently analyzing the interaction relationship between data of different subjects on the multiple data sets, an analysis result for the analysis target is obtained, including: Step S131: grouping the multiple data sets to obtain multiple data groups.

[0044] In specific implementation, multiple data sets can be manually grouped based on experience; alternatively, clustering algorithms such as K-Means and Hierarchical Clustering can be used to automatically group multiple data sets according to their similarity. For example, clustering algorithms can be used to automatically divide data sets that contain highly similar data into the same data group, thereby achieving efficient data classification.

[0045] Step S132: Obtain analysis results for the analysis target by independently analyzing the interaction relationship between data of different topics on the multiple data groups.

[0046] In specific implementation, each data group is analyzed independently to capture the interaction relationship between the data of different subjects in the data group, and then the data of different subjects in the data group are comprehensively analyzed based on the interaction relationship. For example, the reliability of the data of different subjects in the data group can be verified based on the interaction relationship between the data of different subjects (such as positive correlation or negative correlation), and then the erroneous data can be filtered out in time to avoid its impact on the final analysis results.

[0047] In this embodiment, by clustering the data sets into data groups, the number of executions of independent analysis can be reduced, and the richness of data resources for different topics targeted by independent analysis can be further improved, thereby improving the efficiency and accuracy of data analysis.

[0048] Optionally, in step S132, by independently analyzing the interaction relationship between data of different subjects on the multiple data groups, an analysis result for the analysis target is obtained, including: Step S21: For each of the data groups, an independent analysis of the interaction relationship between data of different subjects is performed on the data group.

[0049] In specific implementations, each data group can be input into a pre-trained machine learning model respectively, so that the interaction relationship between data of different topics in each data group can be captured by the machine learning model.

[0050] Step S22: for each of the data groups, according to the analyzed interaction relationship between the data of different topics, data analysis on different topics is performed on the data group respectively.

[0051] In the specific implementation, for each topic, the data of the topic and the data of other topics that interact with it are selected from the data group, and then data analysis of the topic is performed based on the selected data, thereby improving the accuracy of the data analysis results for each topic.

[0052] Step S23: determining the data analysis results obtained for each of the plurality of data groups as analysis results for the analysis target.

[0053] In specific implementation, when the analysis target includes the different topics, that is, the final analysis result is the sum of the analysis results of the different topics, the analysis results of the different topics can be summarized according to the value of the identifier, thereby obtaining a multidimensional analysis result of different objects (such as different user groups) identified by the identifier. Subsequently, more refined business strategy adjustments and optimizations can be performed based on the multidimensional analysis results of different objects (such as different user groups).

[0054] For example, when the multidimensional analysis results of different objects include information such as user behavior patterns (such as shopping behavior habits), health data (such as health status), social media interaction (such as activity, social network influence) of different user groups, the multidimensional analysis results can assist in optimizing business strategies: for example, assisting users to formulate more precise marketing strategies, or assist in health interventions: for example, assisting users to determine personalized health advice and risk warnings, or provide research support: for example, providing in-depth data support for social science or market research, revealing potential causal relationships, etc.

[0055] As a possible implementation, in step S22, according to the interaction relationship between the data of different topics analyzed, the data groups are respectively subjected to data analysis for different topics, including: Step S221: extracting characteristic data corresponding to each of the different topics from the data group.

[0056] In specific implementation, feature data that can reflect specific properties under different topics can be extracted from the data group based on experience or by using a pre-trained neural network model.

[0057] For example, for the topic of user behavior pattern recognition, data such as purchase behavior sequence, shopping records, purchase frequency, and product preference can be extracted from the topic-related data of the data group (such as e-commerce data) as feature data corresponding to the user behavior pattern recognition.

[0058] For another example, with respect to the theme of health risk assessment, data such as amount of exercise and heart rate fluctuations can be extracted from the theme-related data of the data group (such as health indicator data) as feature data corresponding to the health risk assessment.

[0059] Step S222: Based on the analyzed interaction relationship between the data of different themes, use the preset analysis methods corresponding to the different themes to perform data analysis on the characteristic data corresponding to the different themes.

[0060] Among them, the preset analysis method suitable for the different topics can be determined based on experience or through experiments from analysis methods such as principal component analysis (PCA), linear regression, decision tree, etc.

[0061] In specific implementation, for each theme, the data of the theme and the data of other themes that interact with the theme are selected from the data group. Then, based on the selected data, a preset analysis method suitable for the theme is used to perform data analysis on the theme to obtain the data analysis result of the theme.

[0062] In this embodiment, the data of different topics in the data group are analyzed independently according to the topics and interaction relationships, which can reduce the interference of irrelevant data. By using analysis methods suitable for different topics, the characteristics of the data associated with each topic can be captured more accurately, thereby improving the data analysis accuracy of each topic.

[0063] As a possible implementation, in step S131, the multiple data sets are grouped to obtain multiple data groups, including: Step S31: using a clustering algorithm, the plurality of data sets are grouped according to the degree of feature similarity between objects of different categories associated with the data sets to obtain the plurality of data groups.

[0064] In a specific implementation, the multiple data sets may be grouped according to the degree of feature similarity between the associated objects of different classifications by using a clustering algorithm such as K-means or hierarchical clustering.

[0065] Taking K-means as an example, the position of each data set is determined according to the user characteristics (such as age, city or whether or not the user is a member, etc.) of the user groups (i.e., objects of different categories) associated with each data set. The closer the positions of two data sets are, the higher the similarity of the user characteristics of the user groups associated with the two data sets.

[0066] K data sets are randomly selected as the initial cluster centers, and then the distance between each data set and each cluster center is calculated, and each data set is assigned to the cluster center closest to it. When each data set is assigned to the cluster group with the closest cluster center, the center position of each cluster group is used as the new cluster center, and the data sets are reallocated. This process is repeated until the termination condition is met (for example, the cluster center no longer changes), and each cluster group obtained is used as a data group.

[0067] In this embodiment, considering that data analysis is usually performed on the properties of specific objects (such as specific users, specific devices or specific events, etc.), the embodiment of the present application uses a clustering algorithm to group multiple data sets based on the commonalities between objects of different categories, thereby controlling the analysis scope of independent analysis to the data group associated with multiple objects with high feature similarity. This can improve the richness of the data resources targeted by the independent analysis and reduce the interference of irrelevant data (such as data of other objects with lower similarity), thereby achieving improved analysis accuracy.

[0068] Optionally, the accuracy of the analysis results obtained based on the data group is evaluated. If the accuracy is low, the multiple data sets are regrouped by using other clustering algorithms or adjusting the parameters of the clustering algorithm (such as adjusting the number of cluster centers), and the accuracy of the analysis results is evaluated again based on the grouped data groups, until a clustering algorithm or clustering algorithm parameters suitable for the current data analysis scenario are found, which can further improve the accuracy of the data analysis.

[0069] As a possible implementation, the features of an object associated with a single data set are described by a feature vector, and the feature vector is determined by the following steps: Step S41: preprocessing the data set, wherein the preprocessing includes at least one of data cleaning and standardization.

[0070] In specific implementation, the data set can be checked for consistency (for example, whether the value of the variable exceeds the normal value range) or data cleaning such as missing value filling can be performed; and / or, the data set can be standardized using standardization algorithms such as deviation standardization or z-score, thereby improving data quality and helping to improve the accuracy of data analysis.

[0071] Step S42: extracting the multidimensional features of the objects associated with the data set from the preprocessed data set to obtain the feature vector.

[0072] In specific implementation, grouping based on the similarity of feature vectors composed of multi-dimensional features can more comprehensively and accurately analyze the similarity of features presented by different classified objects on different topics, thereby avoiding interference caused by data associated with objects with lower actual feature similarity in subsequent data analysis, thereby further improving the accuracy of data analysis.

[0073] It should be noted that, for the method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited by the described order of actions, because according to the embodiments of the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present application.

[0074] Figure 3 : is a schematic diagram of the structure of a cross-topic data analysis device in an embodiment of the present application, the device comprising: A data collection module is used to collect data on different topics under one or more fields related to the analysis target; A data decoupling module, used for splitting the data of different subjects into multiple data sets according to a preset identifier, wherein the identifier is independent of any single subject, and each of the data sets includes: in the data of different subjects, partial data of multiple subjects associated with the same value of the identifier; The data analysis module is used to obtain analysis results for the analysis target by independently analyzing the interaction relationship between data of different subjects on the multiple data sets.

[0075] By adopting the technical solution of the embodiment of the present application, the data of multiple topics are split and associated using identifiers independent of any single topic, thereby decoupling and reorganizing the data within each topic based on the correlation between the various parts of the data within different topics, thereby providing rich data resources of multiple topics for data analysis through the obtained data set; then, by performing independent cross-topic analysis of multiple data sets, interference between data sets can be avoided, thereby achieving a more refined cross-topic data analysis process, which can greatly improve the accuracy of the final analysis results.

[0076] Optionally, the data analysis module is further configured to perform the following steps: Grouping the multiple data sets to obtain multiple data groups; By independently analyzing the interaction relationship between data of different topics on the multiple data groups, analysis results targeting the analysis target are obtained.

[0077] Optionally, the data analysis module is further configured to perform the following steps: For each of the data groups, performing independent analysis on the interaction relationship between data of different subjects on the data group; For each of the data groups, according to the interaction relationship between the data of different topics analyzed, data analysis on different topics is performed on the data groups respectively; The data analysis results obtained for each of the plurality of data groups are determined as analysis results for the analysis target.

[0078] Optionally, the data analysis module is further configured to perform the following steps: Extracting characteristic data corresponding to each of the different topics from the data group; According to the analyzed interaction relationship between the data of different topics, the preset analysis methods corresponding to the different topics are used to perform data analysis on the characteristic data corresponding to the different topics.

[0079] Optionally, the data analysis module is further configured to perform the following steps: The plurality of data sets are grouped by a clustering algorithm according to the degree of feature similarity between objects of different categories associated with the data sets to obtain the plurality of data groups.

[0080] Optionally, the features of the objects associated with a single data set are described by feature vectors; and the data analysis module is further configured to perform the following steps: Preprocessing the data set, wherein the preprocessing includes at least one of data cleaning and standardization; From the preprocessed data set, multidimensional features of the objects associated with the data set are extracted to obtain the feature vector.

[0081] Optionally, the device further comprises: An object extraction module is used to extract target objects from the data of different subjects, wherein the target objects include: the same or related objects associated with the data of multiple subjects; The identifier generation module is used to generate identifiers with different values ​​for different categories of the target object, and store the identifiers with different values ​​in a database.

[0082] It should be noted that the device embodiment is similar to the method embodiment, so the description is relatively simple, and the relevant parts can be referred to the method embodiment.

[0083] The present application also provides an electronic device, such as Figure 4As shown, it includes a processor 401, a communication interface 402, a memory 403 and a communication bus 404, wherein the processor 401, the communication interface 402, and the memory 403 communicate with each other through the communication bus 404. Memory 403, used for storing computer programs; The processor 401 is used to execute the program stored in the memory 403, such as the steps of the cross-topic data analysis method disclosed in the embodiment of the present application.

[0084] The communication bus mentioned in the above terminal can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0085] The communication interface is used for communication between the above terminal and other devices.

[0086] The memory may include a random access memory (RAM) or a non-volatile memory (non-volatile memory), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0087] The above-mentioned processor can be a general-purpose processor, including a central processing unit, a network processor (Network Processor, NP), etc.; it can also be a digital signal processor (Digital Signal Processing, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0088] The embodiment of the present application also provides a computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the cross-topic data analysis method disclosed in the embodiment of the present application.

[0089] The embodiments of the present application also provide a computer program product, including a computer program / instruction, which, when executed by a processor, implements the cross-subject data analysis method disclosed in the embodiments of the present application.

[0090] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive Solid State Disk (SSD)), etc.

[0091] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0092] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0093] The above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application are included in the protection scope of the present application.

Claims

1. A cross-subject data analysis method, characterized in that: The method comprises: Collect data on different topics under one or more areas related to the analysis objectives; According to a preset identifier, the data of the different subjects are divided into multiple data sets, wherein the identifier is independent of any single subject, and each of the data sets includes: in the data of the different subjects, partial data of multiple subjects associated with the same value of the identifier; By independently analyzing the interaction relationship between data of different subjects on the multiple data sets, analysis results targeting the analysis target are obtained.

2. The method according to claim 1, characterized in that The analysis results for the analysis target are obtained by independently analyzing the interaction relationship between data of different subjects on the multiple data sets, including: Grouping the multiple data sets to obtain multiple data groups; By independently analyzing the interaction relationship between data of different topics on the multiple data groups, analysis results targeting the analysis target are obtained.

3. The method according to claim 2, characterized in that The analysis results for the analysis target are obtained by independently analyzing the interaction relationship between data of different subjects on the multiple data groups, including: For each of the data groups, performing independent analysis on the interaction relationship between data of different subjects on the data group; For each of the data groups, according to the interaction relationship between the data of different topics analyzed, data analysis on different topics is performed on the data groups respectively; The data analysis results obtained for each of the plurality of data groups are determined as analysis results for the analysis target.

4. The method according to claim 3, characterized in that: According to the interaction relationship between the data of different themes analyzed, the data groups are respectively subjected to data analysis for different themes, including: Extracting characteristic data corresponding to each of the different topics from the data group; According to the analyzed interaction relationship between the data of different topics, the preset analysis methods corresponding to the different topics are used to perform data analysis on the characteristic data corresponding to the different topics.

5. The method according to claim 2, characterized in that: The step of grouping the multiple data sets to obtain multiple data groups includes: The plurality of data sets are grouped by a clustering algorithm according to the degree of feature similarity between objects of different categories associated with the data sets to obtain the plurality of data groups.

6. The method according to claim 5, characterized in that The characteristics of an object associated with a single data set are described by a feature vector, and the feature vector is determined by the following steps: Preprocessing the data set, wherein the preprocessing includes at least one of data cleaning and standardization; From the preprocessed data set, multidimensional features of the objects associated with the data set are extracted to obtain the feature vector.

7. The method according to any one of claims 1 to 6, characterized in that: Before splitting the data of different topics into multiple data sets according to the preset identifiers, the method further includes: Extracting target objects from the data of different subjects, wherein the target objects include: the same or related objects associated with the data of multiple subjects; Identifiers with different values ​​are generated for different categories of the target object, and the identifiers with different values ​​are stored in a database.

8. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, for implementing the cross-subject data analysis method as described in any one of claims 1 to 7 when executing a program stored in a memory.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the cross-subject data analysis method according to any one of claims 1 to 7 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the cross-subject data analysis method according to any one of claims 1 to 7 is implemented.

Images