Enterprise multi-modal data fusion visualization method and system

Abstract

The invention relates to the technical field of data processing, in particular to an enterprise multi-modal data fusion visualization method and system, and the method comprises the steps: obtaining information data of each dimension under each type of business data in enterprise multi-modal data; obtaining an association feature factor between each dimension and other dimensions under each type of business data, and integrating the called times of the information data and the time information responded by the called request to obtain a visual attention degree of each dimension under each type of business data; obtaining the degree of suction of different visualization modes in the historical record, and obtaining the degree of membership according to the information data of each dimension under each kind of business data contained in each visualization mode in the historical record; combining the visual attention degree, the membership degree and the suction degree to obtain the priority degree of the information data for each visual mode; and determining an optimal visualization mode of the information data according to the priority degree. According to the method, the visualization mode with a relatively good enterprise data visualization effect can be adaptively determined.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for enterprise multimodal data fusion visualization. Background Art

[0002] In modern enterprise management, the ability to drive decision-making with data has become an important part of enterprise competitiveness. Enterprise operations involve multiple data sources, including sales data, product data, financial data, etc., which usually have different modalities. Enterprise multimodal data fusion visualization generally refers to collecting data from various data sources, and performing preliminary cleaning and integration, and then using data analysis methods to perform data mining and statistical analysis. The processed data is stored in the database, and finally displayed using different visualization methods.

[0003] In existing methods, for multiple categories of enterprise data in enterprise operations, different categories of enterprise data are usually displayed using fixed visualization methods based on experience or historical records. This method does not pay attention to the correlation characteristics between data and the display effects of different visualization methods. Although it improves the efficiency of data visualization to a certain extent, the visualization effect of enterprise data is poor. Summary of the invention

[0004] In order to solve the technical problem that the existing methods have poor visualization effects on various types of enterprise data, the purpose of the present invention is to provide an enterprise multimodal data fusion visualization method and system. The technical solutions adopted are as follows:

[0005] In a first aspect, the present invention provides a method for enterprise multimodal data fusion visualization, the method comprising:

[0006] Obtain information data of each dimension under each type of business data in the enterprise's multimodal data, as well as the number of times the information data is called and the time information of the response to the called request;

[0007] According to the difference between the change trend and the degree of change of the information data of each dimension under each type of business data and the information data of other dimensions of the same type of business data, the correlation characteristic factor between each dimension under each type of business data and other dimensions is obtained;

[0008] According to the correlation characteristic factors between each dimension under each type of business data and all other dimensions under the same type of business data, the data distribution of the number of times the information data is called and the time information of the response to the called request is combined to obtain the visual attention of each dimension under each type of business data;

[0009] Obtain the eye-catching degree of different visualization methods in the historical records, and obtain the membership degree of the information data in each visualization method according to the distribution and time of the information data of each dimension under each business data contained in each visualization method in the historical records;

[0010] The priority level of the information data for each visualization method is obtained by combining the visualization attention level, the membership level of the information data in the corresponding dimension in each visualization method, and the eye-catching level; and the optimal visualization method for the information data is determined according to the priority level.

[0011] Preferably, obtaining the correlation characteristic factor between each dimension under each type of business data and other dimensions according to the difference between the change trend and the difference between the change degree of the information data of each dimension under each type of business data and the information data of other dimensions of the same type of business data specifically includes:

[0012] The information data of any dimension under any type of business data is used as the information data of the target dimension, and the information data of other dimensions under the same type of business data as the information data of the target dimension is used as the information data of the reference dimension;

[0013] determining a first difference factor between the target dimension and each reference dimension based on a difference between a data fluctuation degree of the information data of the target dimension and a data fluctuation degree of the information data of each reference dimension;

[0014] Determining a second difference factor between the target dimension and each reference dimension according to a difference between a change trend of the information data of the target dimension and a change trend of the information data of each reference dimension;

[0015] According to the first difference factor and the second difference factor, an associated characteristic factor between the target dimension and each reference dimension is obtained, and the first difference factor and the second difference factor are both negatively correlated with the associated characteristic factor.

[0016] Preferably, determining the second difference factor between the target dimension and each reference dimension according to the difference between the change trend of the information data of the target dimension and the change trend of the information data of each reference dimension specifically includes:

[0017] Based on the distribution of all information data of the target dimension, each first extreme value point of the target dimension is determined; for any reference dimension, based on the distribution of all information data of the reference dimension, each second extreme value point of the reference dimension is determined;

[0018] The slope of the line between each two adjacent first extreme value points of the target dimension is calculated and recorded as the first slope value of the target dimension; the slope of the line between each two adjacent second extreme value points of the reference dimension is calculated and recorded as the second slope value of the reference dimension;

[0019] A second difference factor between the target dimension and any one of the reference dimensions is determined based on an average value of the difference between each first slope value of the target dimension and each second slope value corresponding to the reference dimension.

[0020] Preferably, the visualization attention of each dimension under each type of business data is obtained based on the correlation characteristic factor between each dimension under each type of business data and all other dimensions under the same type of business data, by combining the data distribution of the number of times the information data is called and the time information of the response to the called request, specifically including:

[0021] According to the data distribution of the number of times all information data of the target dimension are called and the time information of the response to the called request, the data call factor of the target dimension is obtained;

[0022] Obtain the average value of the correlation characteristic factors between the target dimension and all reference dimensions to obtain the data correlation factor of the target dimension;

[0023] According to the data calling factor and the data association factor of the target dimension, the visualization attention of the target dimension is obtained; the data calling factor is positively correlated with the visualization attention, and the data association factor is negatively correlated with the visualization attention.

[0024] Preferably, obtaining the data call factor of the target dimension according to the data distribution of the number of times all information data of the target dimension are called and the time information of the response to the called request specifically includes:

[0025] The time information of the call request response of the information data of the target dimension includes the time interval between the request and the request response each time all the information data of the target dimension are called; the time characteristic value of the target dimension is obtained by calculating the average value of all time intervals corresponding to the information data of the target dimension;

[0026] Based on the ratio between the sum of the number of times all information data of the target dimension are called and the time characteristic value, the data calling factor of the target dimension is determined, and the data calling factor is a normalized value.

[0027] Preferably, obtaining the degree of membership of the information data in each visualization mode according to the distribution and time of the information data in each dimension under each business data contained in each visualization mode in the historical records specifically includes:

[0028] For any visualization method, the total amount of information data of the target dimension in the information data displayed by the visualization method is obtained and recorded as the quantity characteristic value;

[0029] Obtaining the mean of the time intervals between two adjacent pieces of information data of the target dimension displayed in the information data displayed in the visualization mode, and recording the interval characteristic value;

[0030] The product of the proportion of the quantity characteristic value and the negative correlation coefficient of the interval characteristic value is used as the membership degree of the information data of the target dimension in the visualization method.

[0031] Preferably, combining the visualization attention degree with the membership degree of the information data of the corresponding dimension in each visualization mode and the eye-catching degree to obtain the priority degree of the information data for each visualization mode specifically includes:

[0032] For any visualization method, the L2 norm between the visualization attention of the target dimension and the membership of the information data of the target dimension in the visualization method is calculated, and the normalized result of the product of the L2 norm and the eye-catching degree of the visualization method is used as the priority degree of the information data of the target dimension for the visualization method.

[0033] Preferably, obtaining the eye-catchingness of different visualization methods in the historical records specifically includes:

[0034] A neural network is used to obtain the eye-catching degree of each visualization method. The neural network is trained by artificial labeling. The value range of the eye-catching degree is (0, 1).

[0035] Preferably, the business data includes sales data, product data and financial data; the product data includes product sales quantity data, product sales data and product inventory data; each type of business data contains one type of data corresponding to one dimension of information data.

[0036] In a second aspect, the present invention provides an enterprise multimodal data fusion visualization system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the computer program implements the steps of an enterprise multimodal data fusion visualization method when executed by the processor.

[0037] The embodiments of the present invention have at least the following beneficial effects:

[0038] The present invention obtains information data of various types, and obtains the number of times and time information of information data being called, providing a data basis for subsequent feature analysis. First, the correlation characteristic factors between the information data of two dimensions are quantitatively characterized for the differences in two aspects between information data of different dimensions under the same type of business data. The correlation characteristic factors reflect the correlation between each dimension and other dimensions from the differences in the trend of information data change and the degree of information data change. Then, the visualization attention is determined by comprehensively considering the correlation characteristic factors and the distribution of various aspects of information data being called. The visualization attention comprehensively characterizes the importance of information data of the corresponding dimension, as well as the priority and attention of visualization from the two aspects of correlation and data usage. Further, the eye-catching degree of each visualization method is quantified to reflect the quality and popularity of the visualization effect of the corresponding visualization method, and the distribution and time of the information data of each dimension in the visualization method display results in the historical records are analyzed to determine the membership, which characterizes the membership degree of the information data of each dimension using a certain visualization method for visualization display, and the quality of the effect. Finally, by combining the feature analysis results of these aspects, we can determine the priority of displaying the information data of each dimension using a certain visualization method, so that the best visualization method can be adaptively determined for each type of data. The visualization effect is better, and the more appropriate visualization form can more intuitively display the characteristics and trends of the data, helping decision makers make more scientific and reasonable decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 It is a flowchart of the steps of a method for enterprise multimodal data fusion visualization provided by the present invention;

[0041] Figure 2 It is a flowchart of the steps of the method for obtaining the correlation characteristic factor between the target dimension and the reference dimension provided by the present invention;

[0042] Figure 3 is a schematic diagram of a first fitting curve of product sales quantity data provided by the present invention;

[0043] Figure 4 is a schematic diagram of a second fitting curve of product sales data provided by the present invention;

[0044] Figure 5 It is a flowchart of the steps of the method for obtaining visual attention of the target dimension provided by the present invention. DETAILED DESCRIPTION

[0045] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the enterprise multimodal data fusion visualization method and system proposed by the present invention, its specific implementation method, structure, features and effects are described in detail as follows in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0046] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0047] The specific scheme of the enterprise multimodal data fusion visualization method and system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0048] See also Figure 1 , which shows a flowchart of a method for enterprise multimodal data fusion visualization provided by an embodiment of the present invention, the method comprising the following steps:

[0049] Step S100, obtaining information data of each dimension under each type of business data in the enterprise multimodal data, as well as the number of times the information data is called and the time information of the response to the called request.

[0050] In enterprise multimodal data, there are many types of business data. When performing numerical analysis on various types of business data of an enterprise, visual display can more intuitively analyze the changing trends of the data. Different types of business data require different visualization methods in some scenarios, rather than fixed visualization methods, to ensure the visualization effect of enterprise data.

[0051] In this embodiment, data feature analysis is mainly performed on the information data of multiple dimensions contained in each type of business data, as well as visualization effect analysis corresponding to the visualization method, to provide a data basis for the subsequent data analysis process. In this embodiment, the business data includes sales data, product data and financial data, and each type of business data contains information data of multiple dimensions, which is used to display the information related to the enterprise business in more detail.

[0052] For example, product data includes product sales quantity data, product sales data, and product inventory data. Sales data includes sales order quantity, sales order amount, and sales price. Financial data includes profit data, cost data, and asset data. Under each type of business data, one type of data corresponds to one dimension of information data, that is, under the product data type, product sales quantity data belongs to one dimension of information data.

[0053] It should be noted that the business data of the enterprise in this embodiment needs to be obtained under the premise of complying with relevant industry regulations and commercial privacy, and the data obtained is data within a certain time period in the database, that is, the data has a certain time series, and data collection is performed at fixed time intervals. The implementer can set the length of the time period according to the specific implementation scenario, such as 30 days, that is, in this embodiment, one information data can be collected every day. At the same time, in order to prevent the dimension problem from affecting the data analysis results when performing data analysis between different data, this embodiment performs data preprocessing operations on each collected information data, wherein the preprocessing operation may include standardization processing, and the method of standardization processing is a well-known technology and will not be introduced in detail here.

[0054] Furthermore, in the process of enterprise data visualization, different dimensions of information data under different types of business data should have different response priorities. Information data with higher priorities will occupy a larger proportion in the actual access and call process, and thus this type of information data is relatively more important. Therefore, it is also necessary to analyze the situation of information data being called, which can reflect the usage of this type of information data.

[0055] Specifically, in this embodiment, the information of all information data being called under each dimension is obtained within a certain time period, that is, the number of times the information data is called and the time information of the response to the called request, specifically including the total number of times all information data is called under a certain dimension and the time interval between the request and the response to the request each time all information data under a certain dimension is called. For example, the total number of times the product sales quantity data under the product data dimension is called, and the time interval between the request and the response to the request each time it is called. It can be understood that querying, reading and writing data on a certain type of data can all be regarded as a record of the use of this type of data, that is, they can all be recorded as the number of call operations, and the call data can be directly queried through the enterprise's system access data records and log data.

[0056] Step S200, obtaining the correlation characteristic factor between each dimension under each type of business data and other dimensions according to the difference between the change trend and the difference between the change degree of the information data of each dimension under each type of business data and the information data of other dimensions of the same type of business data.

[0057] In the process of visualization priority analysis, there may be multiple data types with the same priority, and there may be certain correlations between the data. When the correlation between different types of data is high, it means that the phenomena reflected by a certain type of data and other types of data are closer or similar, which will in turn overwhelm the importance of these types of data. That is, compared with data with lower correlation, the visualization importance or priority of this type of data is lower. Therefore, it is necessary to first analyze the data correlation between each type of data and other types of data.

[0058] In this embodiment, any data type is used as an example for explanation, that is, the information data of any dimension under any type of business data is used as the information data of the target dimension, and the information data of other dimensions under the same type of business data as the information data of the target dimension is used as the information data of the reference dimension. By analyzing the data differences of the target dimension and each reference dimension in two aspects, the associated characteristic factors of the target dimension and each reference dimension are determined, wherein the first aspect refers to the difference between the change trends of the information data of the target dimension and the reference dimension, and the second aspect refers to the difference between the change degrees of the information data of the target dimension and the reference dimension.

[0059] It can be understood that in this embodiment, product data is used as an example for explanation. For example, the dimension where the product sales quantity data under the product data is located is used as the target dimension, and the dimensions where other types of data under the product data except the product sales quantity data are located are all reference dimensions.

[0060] In this embodiment, if Figure 2 As shown, the method for obtaining the correlation characteristic factors between the target dimension and the reference dimension can be implemented by steps S201 to S203.

[0061] Step S201 : determining a first difference factor between the target dimension and each reference dimension based on the difference between the data fluctuation degree of the information data of the target dimension and the data fluctuation degree of the information data of each reference dimension.

[0062] In this embodiment, any reference dimension is taken as an example for description. For example, the i-th dimension under any type of business data is taken as the target dimension, and the v-th dimension under the type of business data except the i-th dimension is taken as the any reference dimension.

[0063] Each acquisition time point under the target dimension corresponds to an information data. Similarly, each acquisition time point under the reference dimension corresponds to an information data. By calculating the variance of all information data under the target dimension, the data fluctuation degree of the information data under the target dimension can be reflected. Similarly, by calculating the variance of all information data under the reference dimension, the data fluctuation degree of the information data under the reference dimension can be reflected. That is, this embodiment uses variance to characterize the data fluctuation degree of a set of data. In other embodiments, the implementer can also select other parameters for characterization according to the specific implementation scenario, such as standard deviation.

[0064] Furthermore, by comparing the difference in data fluctuation between the target dimension and the reference dimension, the correlation characteristics between the target dimension and the reference dimension can be quantitatively characterized. The greater the difference in data fluctuation between the two, the smaller the correlation characteristics between the two.

[0065] Specifically, the absolute value of the difference between the variance of all information data of the target dimension and the variance of all information data of the reference dimension is taken as the first difference factor between the target dimension and the reference dimension. That is, the first difference factor characterizes the difference in data fluctuations between the target dimension and the reference dimension. The larger the value, the greater the difference in data fluctuations.

[0066] Step S202: determining a second difference factor between the target dimension and each reference dimension according to the difference between the change trend of the information data of the target dimension and the change trend of the information data of each reference dimension.

[0067] In this embodiment, the slope of the straight line between the local data is used to reflect the local change trend of the information data in the corresponding dimension, and then the difference in the change trend between the information data of the target dimension and the information data of the reference dimension is analyzed to quantitatively characterize the correlation characteristics of the second aspect.

[0068] First, based on the distribution of all information data of the target dimension, each first extreme point of the target dimension is determined; for any reference dimension, based on the distribution of all information data of the reference dimension, each second extreme point of the reference dimension is determined. The method of obtaining the extreme value of a set of data is a well-known technology and will not be introduced in detail here.

[0069] More specifically, a rectangular coordinate system is constructed with the acquisition time point as the horizontal coordinate and the information data of the target dimension at each acquisition time point as the vertical coordinate, and data fitting is performed to obtain a first fitting curve. At the same time, a rectangular coordinate system is constructed with the acquisition time point as the horizontal coordinate and the information data of the reference dimension at each acquisition time point as the vertical coordinate, and data fitting is performed to obtain a second fitting curve. By performing data analysis on the first fitting curve and the second fitting curve respectively, the first extreme point of the target dimension and the second extreme point of the reference dimension can be obtained. Figure 3 , which shows the first fitting curve of product sales quantity data, i represents the i-th dimension of business data, such as Figure 4 , which shows the second fitting curve of product sales data, v represents the vth dimension of business data, i≠v, and i and v belong to the same type of business data.

[0070] Then, the slope of the line between each two adjacent first extreme value points of the target dimension is calculated and recorded as the first slope value of the target dimension; the slope of the line between each two adjacent second extreme value points of the reference dimension is calculated and recorded as the second slope value of the reference dimension. That is, there is a first slope value corresponding to each two adjacent first extreme value points, which can reflect the local change trend of the first fitting curve at this position, and there is a second slope value corresponding to each two adjacent second extreme value points, which can reflect the local change trend of the second fitting curve at this position.

[0071] For example, Figure 3 In, K i,1 It represents the first slope value corresponding to the first local position of the i-th dimension, i.e., the target dimension, under the business data, which is also the first first slope value. It can be seen from the figure that the first slope value reflects the general trend of data changes from 0 to 3 days. Figure 4 Middle K v,1 It represents the second slope value corresponding to the first local position of the vth dimension, i.e., the reference dimension, under the business data, which is also the first second slope value. It can be seen from the figure that the second slope value reflects the general trend of data changes from 0 to 3 days.

[0072] Finally, based on the average value of the difference between each first slope value of the target dimension and each second slope value corresponding to the reference dimension, a second difference factor between the target dimension and any one of the reference dimensions is determined.

[0073] The first slope value reflects the local change trend between the information data under the target dimension, and the second slope value reflects the local change trend between the information data under the reference dimension. By comparing the difference between the first slope value and the second slope value corresponding to the same position, the difference in the local change trend between the target dimension and the reference dimension can be characterized. Then, the data is integrated by averaging, and finally the second difference factor characterizes the difference in the local change trend between the target dimension and the reference dimension. The larger its value, the smaller the correlation feature between the target dimension and the reference dimension.

[0074] It should be noted that when the number of the first slope value and the second slope value is different, the slope value with the least number is used as a control for difference analysis. Figure 3 and Figure 4 The vertical axis has no unit, and the horizontal axis is time, the unit is day.

[0075] Step S203: obtaining, according to the first difference factor and the second difference factor, an associated characteristic factor between the target dimension and each reference dimension, wherein the first difference factor and the second difference factor are both negatively correlated with the associated characteristic factor.

[0076] The first difference factor reflects the difference in fluctuation between the target dimension and the reference dimension, and the second difference factor reflects the difference in change trend between the target dimension and the reference dimension. By combining the data difference characteristics of the two aspects, the correlation characteristic factor between the target dimension and the reference dimension can be quantitatively characterized. That is, the greater the difference, the smaller the correlation characteristic factor. Therefore, the difference analysis results should be negatively correlated.

[0077] As a specific example, the calculation formula of the correlation characteristic factor between the target dimension and a reference dimension can be expressed as:

[0078]

[0079] Among them, W i,v K represents the correlation characteristic factor between the target dimension and the vth reference dimension, i represents the ith dimension under any type of business data, v represents the vth reference dimension of the same type of business data as the ith dimension, i≠v; i,n Indicates the nth first slope value under the target dimension, K v,n represents the nth first slope value under the reference dimension, N1 represents the minimum number of slope values ​​in the target dimension and the reference dimension, σ i Indicates the degree of data fluctuation of the target dimension, that is, the variance of all information data under the target dimension; σv It represents the degree of data fluctuation in the reference dimension, that is, the variance of all information data under the reference dimension. exp represents an exponential function with the natural constant e as the base.

[0080] |σ i -σ v | represents the first difference factor, It represents the second difference factor. The product of the first difference factor and the second difference factor is negatively correlated in the form of negative exponential power exp(-x). The correlation characteristic factor characterizes the data correlation characteristics between the information data corresponding to the target dimension and the reference dimension. The larger its value is, the more similar the fitting curves corresponding to the target dimension and the reference dimension are not only in the change trend, but also in the degree of fluctuation. There is a high similarity, which means that the correlation between the target dimension and the reference dimension is greater.

[0081] Step S300, based on the correlation characteristic factors between each dimension under each type of business data and all other dimensions under the same type of business data, the data distribution of the number of times the information data is called and the time information of the response to the called request is comprehensively considered to obtain the visual attention of each dimension under each type of business data.

[0082] During the visualization priority analysis of information data of each dimension under each type of business data, the more times the information is called, the more frequently the information data of this type is used, and further, the greater the importance of the information data of this type, which to a certain extent reflects the high urgency of visualizing this type of information data. The larger the value of the correlation characteristic factor between the target dimension and other dimensions, the greater the correlation characteristics of this type of information data with other dimensions, and further, the stronger the substitutability of this type, that is, when performing enterprise data analysis, other dimensions with greater correlation can replace the data of this dimension for feature analysis, and further, the lower the importance, the lower the urgency of visualizing this type of information data. Based on this, this embodiment quantitatively represents the visualization attention of information data under the corresponding dimension by combining the correlation characteristic factor and the information distribution of the information data of the corresponding dimension being called.

[0083] In this embodiment, information data of any dimension under any type of business data is still used as an example for description. Figure 5 As shown, the method for acquiring the visual attention of the target dimension can be implemented by steps S301 to S303.

[0084] Step S301, obtaining the data call factor of the target dimension according to the data distribution of the number of times all information data of the target dimension are called and the time information of the response to the called request.

[0085] The more frequently all information data under the target dimension are called, the more frequently the data of the corresponding type in the target dimension is retrieved, analyzed, and used by relevant personnel of the enterprise, which further indicates that the data of the corresponding type in the target dimension is of high importance. Among them, the information on the situation of all information data of the target dimension being called includes two aspects, which have been introduced in step S100. The first aspect includes the number of calls, specifically the total number of calls of all information data of the target dimension. The second aspect includes the time information of the response to the called request, specifically the time interval between the request and the response to the request each time all information data of the target dimension is called.

[0086] Under the target dimension, each time the information data is called, there is a time interval. By comprehensively analyzing the time intervals of all calling operations, the data call response of the corresponding type of the target dimension can be reflected, that is, the average value of all time intervals corresponding to the information data of the target dimension is calculated to obtain the time characteristic value of the target dimension; this time characteristic value reflects the frequency of use of the corresponding type of data in the target dimension.

[0087] Furthermore, based on the ratio between the sum of the number of times all the information data of the target dimension are called and the time characteristic value, the data call factor of the target dimension is determined, and the data call factor is a normalized value. In this embodiment, the result after normalizing the ratio is used as the data call factor of the target dimension, wherein the normalization method is a well-known technology and will not be introduced in detail here. The data call factor characterizes the situation in which the information data of the target dimension is called and the frequency of data use. The more times the information data of the target dimension is called, the shorter the call response time, and the larger the value of the corresponding data call factor, indicating that the data is used more frequently or is called more frequently.

[0088] Step S302: Obtain the average value of the correlation characteristic factors between the target dimension and all reference dimensions to obtain the data correlation factor of the target dimension.

[0089] There is a correlation characteristic factor between the target dimension and each reference dimension, which is used to reflect the correlation characteristics of the information data between the target dimension and each reference dimension. The average value of all the correlation characteristic factors corresponding to the target dimension is taken as the data correlation factor of the target dimension. The data correlation factor integrates the correlation characteristics between the target dimension and all other reference dimensions, and reflects the correlation size of the information data of the target dimension under the overall comprehensive situation.

[0090] Step S303, obtaining the visualization attention degree of the target dimension according to the data calling factor and the data association factor of the target dimension.

[0091] Specifically, the data call factor of the target dimension is positively correlated with the visualization attention, and the data association factor of the target dimension is negatively correlated with the visualization attention. In this embodiment, the calculation formula of the visualization attention of the target dimension can be expressed as:

[0092]

[0093] Among them, S i represents the visual attention of the target dimension, F i represents the data call factor of the target dimension, C represents the total number of all reference dimensions corresponding to the target dimension, and W i,v It represents the correlation characteristic factor between the target dimension and the vth reference dimension, i represents the i-th dimension under any type of business data, v represents the v-th reference dimension belonging to the same type of business data as the i-th dimension, i≠v; exp() represents an exponential function with the natural constant e as the base.

[0094] Data recall factor F i The larger the value of, the more frequently the data is used. The smaller the value of is, the smaller the correlation characteristics between the information data of the target dimension and the information data of other dimensions are, the smaller the substitutability of the information data of the target dimension is, the greater the importance of the corresponding data is, and thus the visualization attention S of the target dimension is higher. i The larger the value of . The visualization attention of the target dimension represents the attention, importance and priority of the visualization of information data under the target dimension.

[0095] Step S400, obtaining the eye-catching degree of different visualization methods in the historical records, and obtaining the membership degree of the information data in each visualization method according to the distribution and time of the information data of each dimension under each business data contained in each visualization method in the historical records.

[0096] For enterprises, general data visualization methods include line charts, bar charts, pie charts, scatter plots, heat maps, and three-dimensional graphs, etc. Different visualization methods have different display effects and popularity. When using different visualization methods to visualize the information data in business data, enterprise personnel have different concerns and preferences for different display results. In the historical records, there are records of visualization of information data in each dimension under each business data. The distribution of different types of information data under each visualization method can be analyzed, and the display effects and popularity of different visualization methods can be understood through the historical display records. Finally, the degree of visualization of each type of information data using each visualization method can be quantified.

[0097] First, it is necessary to evaluate the visual display effects of different visualization methods. In this embodiment, this is achieved by means of a neural network. Specifically, the eye-catching degree of each visualization method is obtained by using a neural network. The neural network is trained by artificial labeling, and the value range of the eye-catching degree is (0, 1). The eye-catching degree represents the quality of the visualization effect and the popularity of the corresponding visualization method.

[0098] More specifically, in this embodiment, the training information of the neural network specifically includes: feature information of different visualization methods includes bar charts, pie charts, line charts, scatter plots, and heat maps; each feature information is scored by manual scoring, and the higher the score, the more eye-catching the specific data visualization display form is; the neural network is a classification network structure with a five-layer fully connected neural network model, and the loss function is a cross entropy function; the training ratio is set to 7:3, and the eye-catching degree of the visualization display form is trained through the fully connected network model. The input of the classification network is a bar chart, pie chart, line chart, scatter plot, and heat map, and the output is the eye-catching degree corresponding to each visualization method.

[0099] During the classification network training process, the gradient descent method is used for training until the loss function converges, and the classification network training is completed. The value range of eye-catching degree is (0,1). Eye-catching degree represents the effect and popularity of the corresponding visualization method. The higher the eye-catching degree, the more popular the visualization display form is and the better the visualization effect is.

[0100] Then, in the historical records, each visualization method contains a variety of information data. By counting the proportion of information data corresponding to each visualization method, it can be reflected that the amount of data visualized in the historical records using each visualization method. The larger the proportion of data, the more it can be explained that this type of information data in the historical records is more inclined to be displayed using this visualization method.

[0101] Specifically, for any visualization method, the total number of information data of the target dimension in the information data displayed by the visualization method is obtained and recorded as the quantity characteristic value; the average of the time intervals between each two adjacent displays of the information data of the target dimension in the information data displayed by the visualization method is obtained and recorded as the interval characteristic value; it should be noted that the time interval records the time interval between two adjacent visualization displays of the information data of the target dimension in the historical records using the same visualization method. For example, in the historical records, a total of 100 visualization displays were performed using a certain visualization method, among which the second display, the sixth display, and the seventh display all visualized the information data of the target dimension, and a time record was kept for each visualization display, so that the time interval between the second display and the sixth display, and the time interval between the sixth display and the seventh display can be obtained. The average of these two time intervals is the interval characteristic value, which reflects the frequency of visualization of the information number of the target dimension under this visualization display method in terms of time distribution.

[0102] Furthermore, the product of the proportion of the quantity characteristic value and the negative correlation coefficient of the interval characteristic value is used as the membership degree of the information data of the target dimension in the visualization method. As a specific example, taking any visualization method as an example, the calculation formula of the membership degree of the information data of the target dimension in the u-th visualization method can be expressed as:

[0103]

[0104] Among them, Gh u,i represents the membership degree of the information data of the target dimension in the u-th visualization method, G u,i represents the total amount of information data of the target dimension contained in the information data displayed in the u-th visualization method, that is, the quantitative characteristic value; G u It represents the total number of all information data displayed in the u-th visualization method; ΔT represents the mean time interval between each two adjacent information data of the target dimension displayed in the u-th visualization method, that is, the interval characteristic value; i represents the i-th dimension under any type of business data, that is, the target dimension.

[0105] Indicates the proportion of information data of the target dimension in the information data displayed in the u-th visualization method, The larger the value of , the more information data of the target dimension is contained in the information data displayed in the u-th visualization method, and the higher the visualization frequency. This further shows that in the actual enterprise data visualization history, the information data of the target dimension is better visualized using the u-th visualization method, and the acceptance of enterprise personnel is relatively high, and then Ghu,i The larger the value of is, the higher the membership of the information data of the target dimension to the u-th visualization method is.

[0106] The membership degree represents the degree of membership of this type of information data displayed using the u-th visualization method. The larger the value, the greater the membership degree, which further indicates that the effect of displaying this type of information data using the u-th visualization method is better.

[0107] Step S500, combining the visualization attention and the membership of the information data of the corresponding dimension in each visualization method and the eye-catching degree, obtain the priority degree of the information data for each visualization method; determine the best visualization method for the information data according to the priority degree.

[0108] Under different visualization methods, the visualization effects of information data of different dimensions under different types of business data are different. In order to determine the best visualization display method, each visualization method has a certain priority for each type of information data. In order to more accurately evaluate the priority, this embodiment comprehensively considers two aspects of data characteristics and quantifies the tendency of the visualization effect of the visualization method.

[0109] First, by analyzing the degree of visualization attention paid to each type of information data within a period of time, the attention, importance and priority of visualization of this type of data can be reflected. If the correlation of information data within a period of time is poor and it is called more frequently, the higher the importance of this type of information data, the greater the priority of visualization.

[0110] Secondly, by analyzing the proportion and time distribution of data in each different visualization method in the historical records, the connection between the visualization method and each type of information data is constructed, and the membership degree is used to reflect the degree of preference of each type of information data for each visualization method. The higher the membership degree, the better the visualization effect of this type of information data using the corresponding visualization method, and the greater the priority of visualization.

[0111] Based on this, for any visualization method, the L2 norm between the visualization attention of the target dimension and the membership of the information data of the target dimension in the visualization method is calculated, and the normalized result of the product of the L2 norm and the eye-catching degree of the visualization method is used as the priority degree of the information data of the target dimension for the visualization method.

[0112] As a specific example, taking any visualization method as an example, the calculation formula of the priority degree of the information data of the target dimension for the u-th visualization method can be expressed as:

[0113]

[0114] Among them, R i,u Indicates the priority of the target dimension information data for the u-th visualization method, α u represents the eye-catching degree of the u-th visualization method, S i Represents the visual attention of the target dimension, Gh u,i It represents the degree of membership of the information data of the target dimension in the u-th visualization method, and th() represents the hyperbolic tangent function.

[0115] use The features of visualization attention and membership are integrated. When the values ​​of both are large, the priority of the corresponding visualization method is higher. u Taking the feature synthesis as a coefficient, the larger the eye-catching value is, the better the visualization effect of the corresponding visualization method is, which means that the priority of this visualization method for data processing is higher.

[0116] The priority level represents the priority of visualizing information data using the corresponding visualization method. The larger the value is, the better the effect of visualizing the information data using this visualization method is, and the higher the priority is.

[0117] At this point, each dimension of information data under each type of business data corresponds to a priority level between each visualization method, reflecting the effect of processing the corresponding data using the corresponding visualization method. You can perform a screening operation on the visualization method for each type of information data.

[0118] For example, for the information data of the target dimension, the visualization method corresponding to the maximum value of the priority levels corresponding to all visualization methods is used as the optimal visualization method for the information data of the target dimension.

[0119] Visualizing the corresponding type of data according to the best visualization method can achieve the best visualization effect, and to a certain extent can avoid the situation where a large amount of data causes system delays, improve system performance and system visualization. Appropriate visualization methods can help companies understand data more effectively, reveal patterns and trends, and thus support data-driven decision-making.

[0120] The embodiment of the present invention also provides an enterprise multimodal data fusion visualization system, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the computer program implements the steps of an enterprise multimodal data fusion visualization method when executed by the processor. Since an embodiment of an enterprise multimodal data fusion visualization method has been described in detail, the process will not be repeated here.

[0121] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for enterprise multimodal data fusion visualization, characterized in that: The method comprises the following steps: Obtain information data of each dimension under each type of business data in the enterprise's multimodal data, as well as the number of times the information data is called and the time information of the response to the called request; According to the difference between the change trend and the degree of change of the information data of each dimension under each type of business data and the information data of other dimensions of the same type of business data, the correlation characteristic factor between each dimension under each type of business data and other dimensions is obtained; According to the correlation characteristic factors between each dimension under each type of business data and all other dimensions under the same type of business data, the data distribution of the number of times the information data is called and the time information of the response to the called request is combined to obtain the visual attention of each dimension under each type of business data; Obtain the eye-catching degree of different visualization methods in the historical records, and obtain the membership degree of the information data in each visualization method according to the distribution and time of the information data of each dimension under each business data contained in each visualization method in the historical records; The priority level of the information data for each visualization method is obtained by combining the visualization attention level, the membership level of the information data in the corresponding dimension in each visualization method, and the eye-catching level; and the optimal visualization method for the information data is determined according to the priority level.

2. The enterprise multimodal data fusion visualization method according to claim 1 is characterized in that: The method of obtaining the correlation characteristic factors between each dimension and other dimensions under each type of business data according to the difference between the change trends and the difference between the change degrees of the information data of each dimension under each type of business data and the information data of other dimensions of the same type of business data specifically includes: The information data of any dimension under any type of business data is used as the information data of the target dimension, and the information data of other dimensions under the same type of business data as the information data of the target dimension is used as the information data of the reference dimension; determining a first difference factor between the target dimension and each reference dimension based on a difference between a data fluctuation degree of the information data of the target dimension and a data fluctuation degree of the information data of each reference dimension; Determining a second difference factor between the target dimension and each reference dimension according to a difference between a change trend of the information data of the target dimension and a change trend of the information data of each reference dimension; According to the first difference factor and the second difference factor, an associated characteristic factor between the target dimension and each reference dimension is obtained, and the first difference factor and the second difference factor are both negatively correlated with the associated characteristic factor.

3. The enterprise multimodal data fusion visualization method according to claim 2 is characterized in that: The determining, according to the difference between the change trend of the information data of the target dimension and the change trend of the information data of each reference dimension, a second difference factor between the target dimension and each reference dimension specifically includes: Based on the distribution of all information data of the target dimension, each first extreme value point of the target dimension is determined; for any reference dimension, based on the distribution of all information data of the reference dimension, each second extreme value point of the reference dimension is determined; The slope of the line between each two adjacent first extreme value points of the target dimension is calculated and recorded as the first slope value of the target dimension; the slope of the line between each two adjacent second extreme value points of the reference dimension is calculated and recorded as the second slope value of the reference dimension; A second difference factor between the target dimension and any one of the reference dimensions is determined based on an average value of the difference between each first slope value of the target dimension and each second slope value corresponding to the reference dimension.

4. The enterprise multimodal data fusion visualization method according to claim 2 is characterized in that: According to the correlation characteristic factor between each dimension under each type of business data and all other dimensions under the same type of business data, the data distribution of the number of times the information data is called and the time information of the response to the called request is combined to obtain the visualized attention of each dimension under each type of business data, specifically including: According to the data distribution of the number of times all information data of the target dimension are called and the time information of the response to the called request, the data call factor of the target dimension is obtained; Obtain the average value of the correlation characteristic factors between the target dimension and all reference dimensions to obtain the data correlation factor of the target dimension; According to the data calling factor and the data association factor of the target dimension, the visualization attention of the target dimension is obtained; the data calling factor is positively correlated with the visualization attention, and the data association factor is negatively correlated with the visualization attention.

5. The enterprise multimodal data fusion visualization method according to claim 4 is characterized in that: The data calling factor of the target dimension is obtained according to the data distribution of the number of times all information data of the target dimension are called and the time information of the called request response, specifically including: The time information of the call request response of the information data of the target dimension includes the time interval between the request and the request response each time all the information data of the target dimension are called; the time characteristic value of the target dimension is obtained by calculating the average value of all time intervals corresponding to the information data of the target dimension; Based on the ratio between the sum of the number of times all information data of the target dimension are called and the time characteristic value, the data calling factor of the target dimension is determined, and the data calling factor is a normalized value.

6. The enterprise multimodal data fusion visualization method according to claim 2 is characterized in that: The obtaining of the degree of subordination of the information data in each visualization mode according to the distribution and time of the information data in each dimension under each business data in each visualization mode in the historical records specifically includes: For any visualization method, the total amount of information data of the target dimension in the information data displayed by the visualization method is obtained and recorded as the quantity characteristic value; Obtaining the mean of the time intervals between two adjacent pieces of information data of the target dimension displayed in the information data displayed in the visualization mode, and recording the interval characteristic value; The product of the proportion of the quantity characteristic value and the negative correlation coefficient of the interval characteristic value is used as the membership degree of the information data of the target dimension in the visualization method.

7. The enterprise multimodal data fusion visualization method according to claim 2 is characterized in that: The combining of the visualization attention and the membership of the information data of the corresponding dimension in each visualization mode and the eye-catching degree to obtain the priority of the information data for each visualization mode specifically includes: For any visualization method, the L2 norm between the visualization attention of the target dimension and the membership of the information data of the target dimension in the visualization method is calculated, and the normalized result of the product of the L2 norm and the eye-catching degree of the visualization method is used as the priority degree of the information data of the target dimension for the visualization method.

8. The enterprise multimodal data fusion visualization method according to claim 1 is characterized in that: The obtaining of the eye-catchingness of different visualization methods in the historical records specifically includes: A neural network is used to obtain the eye-catching degree of each visualization method. The neural network is trained by artificial labeling. The value range of the eye-catching degree is (0, 1).

9. The enterprise multimodal data fusion visualization method according to claim 1, characterized in that: The business data includes sales data, product data and financial data; the product data includes product sales quantity data, product sales data and product inventory data; each type of business data contains a type of data corresponding to a dimension of information data.

10. An enterprise multimodal data fusion visualization system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the computer program is executed by a processor, the steps of the enterprise multimodal data fusion visualization method as described in any one of claims 1 to 9 are implemented.

Images