Data asset display analysis method and system based on data visualization

Abstract

The application provides a data asset display analysis method and system based on data visualization, and belongs to the technical field of data asset display analysis. First, the original characteristic information of the structural characteristics, the correlation characteristics and the flow characteristics of the data asset is collected, then the visual dimension mapping processing is performed to obtain a visual dimension set containing the structural dimension, the correlation dimension and the flow dimension, a multi-level display template containing the basic display layer, the correlation display layer and the dynamic display layer is constructed according to the visual dimension set, the visual interactive interface is generated by performing the interactive adaptation processing on the multi-level display template according to the user analysis requirement, the operation track data of the user in the visual interactive interface is tracked, the multi-level display template is adjusted and the visual interactive interface is updated based on the operation track data, and the comprehensive, intuitive and interactive display analysis of the data asset is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data asset display analysis, in particular to a data asset display analysis method and system based on data visualization. BACKGROUND

[0002] In today's digital era, data assets have become one of the core resources of enterprises and organizations, and their value is increasingly prominent. In order to fully tap and utilize the value of data assets, it is necessary to effectively display and analyze data assets. At present, the common data asset display and analysis method mainly relies on traditional reports and charts. Although the above-mentioned method can present certain data information, it has many limitations.

[0003] On the one hand, the traditional display method can only present the single characteristics of data assets, and it is difficult to fully display the multi-dimensional characteristics such as structure, association and flow of data assets. For example, when displaying the structure of data assets, it may only present the fields and types of data in the form of a simple table, and it is difficult to clearly display the hierarchical relationship and complex structure between data. When displaying data association, it is also difficult to intuitively present the internal relationship and influence between different data assets. On the other hand, the traditional display method lacks interactivity and dynamics, and users can only passively view the preset reports and charts, and cannot flexibly interact and dynamically adjust the display according to their own analysis needs, which makes it difficult for users to deeply mine the potential information and value behind the data when analyzing data assets, and cannot meet the increasingly complex and changing data analysis needs. SUMMARY

[0004] In view of the above-mentioned problems, in combination with the first aspect of the present application, the embodiments of the present application provide a data asset display analysis method based on data visualization, which comprises:

[0005] Collecting the original characteristic information of the data assets, wherein the original characteristic information includes the structural characteristics, the association characteristics and the flow characteristics of the data assets;

[0006] Performing visual dimension mapping processing on the original characteristic information to obtain a visual dimension set of the data assets, wherein the visual dimension set includes a structure dimension, an association dimension and a flow dimension;

[0007] According to the visual dimension set, a multi-level display template of the data assets is constructed, wherein the multi-level display template includes a basic display layer, an association display layer and a dynamic display layer;

[0008] According to the analysis needs of the user, the multi-level display template is interactively adapted to generate a visual interactive interface of the data assets;

[0009] Tracking operation trajectory data of the user in the visualization interactive interface, adjusting the multi-level display template based on the operation trajectory data, and updating the visualization interactive interface.

[0010] In still another aspect, the embodiment of the present application further provides a data asset display analysis system based on data visualization, comprising a processor, a machine readable storage medium, the machine readable storage medium being connected with the processor, the machine readable storage medium being used for storing programs, instructions or codes, and the processor being used for executing the programs, instructions or codes in the machine readable storage medium to realize the method described above.

[0011] Based on the above aspects, the embodiment of the present application realizes comprehensive extraction and integration of multi-dimensional features of data assets by collecting original feature information such as structural features, correlation features and flow features of data assets and performing visualization dimension mapping processing to obtain a visualization dimension set containing structural dimensions, correlation dimensions and flow dimensions. The multi-level display template constructed according to the visualization dimension set contains a basic display layer, a correlation display layer and a dynamic display layer, which displays data assets from different levels and angles, can meet the diversified analysis needs of users, and enables users to more comprehensively and deeply understand various aspects of data assets. The multi-level display template is interactively adapted according to the analysis needs of users to generate a visualization interactive interface of data assets, enhances the interactivity between users and data assets, and enables users to flexibly view and analyze data according to their own will. The operation trajectory data of the user in the visualization interactive interface is tracked, and the multi-level display template is adjusted and the visualization interactive interface is updated based on the data, which realizes the dynamic and personalized display analysis, can optimize the display effect in real time according to the operation habits and analysis focus of users, and greatly improves the efficiency and accuracy of data asset display analysis. BRIEF DESCRIPTION OF DRAWINGS

[0012] Figure 1 is a flowchart of the method for displaying and analyzing data assets based on data visualization provided by an embodiment of the present application.

[0013] Figure 2 is a schematic diagram of exemplary hardware and software components of the system for displaying and analyzing data assets based on data visualization provided by an embodiment of the present application. DETAILED DESCRIPTION

[0014] The present application will be described in detail below with reference to the accompanying drawings of the specification, Figure 1 is a flowchart of the method for displaying and analyzing data assets based on data visualization provided by an embodiment of the present application, which will be described in detail below.

[0015] Step S110: Collect original feature information of the data asset, which includes structural features, correlation features and flow features of the data asset.

[0016] In this embodiment, the data asset in the enterprise customer relationship management system is taken as the application scenario to collect the original feature information. First, the scope of the data asset needs to be clarified, covering customer basic information dataset, transaction record dataset, customer feedback dataset and other parts. These data assets are stored and transferred in different forms in the enterprise customer relationship management system.

[0017] Step S111: Access the data asset management system, and call the metadata records of all data assets stored in the data asset management system, which include the field composition, format type and storage path of the data asset.

[0018] In the enterprise customer relationship management system, the data asset management system records the metadata information of all data assets. When accessing the data asset management system, a connection needs to be established through the interface provided by the data asset management system. The interface adopts a standard communication protocol to ensure the stability and security of data transmission. After the connection is successful, a call request is sent, which includes the scope of the metadata records to be obtained, i.e. all data assets related to customer relationship management.

[0019] After receiving the request, the data asset management system can retrieve the internally stored metadata records. For the customer basic information dataset, the field composition in the metadata record includes customer name, gender, age, contact information, industry, etc.; the format type is structured data, stored in the form of a table; and the storage path is the customer information table path in the system database. The field composition of the transaction record dataset includes transaction number, customer identifier, transaction time, transaction amount, transaction goods, etc.; the format type is also structured data; and the storage path is the transaction information table path. The field composition of the customer feedback dataset includes feedback number, customer identifier, feedback time, feedback content, feedback type, etc.; the format type includes structured data (such as feedback number and customer identifier) and unstructured data (such as text information of feedback content); and the storage paths correspond to the feedback information table and the text storage directory, respectively.

[0020] Step S112: Extract the hierarchical relationship of field composition, the dependency relationship between fields and the compatible attributes of format type from the metadata records, and integrate to form the structural features of the data asset, which are presented in the form of a hierarchical tree.

[0021] For the customer basic information dataset, the hierarchical relationship of field composition is analyzed from the metadata record. The customer basic information serves as the root node, which is divided into two child nodes of personal information and enterprise information. The personal information child node further contains fields of name, gender, age, contact information, etc. The enterprise information child node contains fields of enterprise name, industry, enterprise size, etc.

[0022] In terms of the dependency relationship between fields, the customer identification field is the key for the association of the customer basic information dataset with other datasets. In the customer basic information dataset, the contact information field depends on the existence of the customer name field, i.e., each customer record must have a customer name, and then the contact information is recorded.

[0023] In terms of the compatibility attribute of format type, the customer name field is in text format, the gender field is in enumeration format (male / female / other), and the age field is in integer format. There are certain compatibility rules between the above formats, such as the integer format of the age field cannot be directly operated with the text format of the name field, but can be text spliced in display.

[0024] After integrating the above information, the structural characteristics are formed in the form of a hierarchical tree. The root node is the customer basic information dataset, which is divided into child nodes of personal information and enterprise information. Each child node further contains corresponding fields as leaf nodes. Meanwhile, the dependency relationship between fields and the format compatibility attribute are marked in the tree structure by specific markers.

[0025] Step S113: Query the association mapping table in the data asset management system to extract the reference relationship, shared relationship, and combination relationship between different data assets, and integrate them into the association characteristics of the data assets. The reference relationship refers to the field calling relationship of one data asset to another data asset, the shared relationship refers to the set of basic fields commonly used by multiple data assets, and the combination relationship refers to the aggregated form of multiple data assets according to the set rules.

[0026] The association mapping table in the data asset management system records the association between data assets. Querying the table shows that the customer basic information dataset and the transaction record dataset have a reference relationship. The customer identification field in the transaction record dataset calls the customer identification field in the customer basic information dataset, and the association query of the two datasets is realized through the customer identification field.

[0027] In terms of the shared relationship, the customer basic information dataset, the transaction record dataset, and the customer feedback dataset share the customer identification field, which is a basic field commonly used by the three datasets for uniquely identifying customers and realizing cross-dataset customer information tracking.

[0028] In terms of combination relationship, the customer value analysis dataset is formed by aggregating the customer basic information dataset, the transaction record dataset and the customer feedback dataset according to a set rule. The aggregation rule is: according to the customer identifier, the related records in the three datasets are matched, the basic information, the total transaction amount, the transaction frequency, the feedback satisfaction and other information of the customer are extracted, and the comprehensive data of the customer value evaluation is formed.

[0029] After integrating these reference relationships, sharing relationships and combination relationships, the association features of the data assets are formed, and are presented in the form of a clear relationship list, with the data set name, association type and specific association content being marked.

[0030] Step S114: Obtain the transmission record of the data asset between different processing nodes, extract the node identifier, transmission time and processing state in the transmission process, generate the flow feature of the data asset, and the node identifier includes the generation node, the processing node and the storage node, and the processing state includes to be processed, in processing and completed.

[0031] The flow process of the customer feedback dataset in the system has a clear transmission record. The generation node is the customer feedback submission module. After the customer submits the feedback information through the customer feedback submission module, the data first enters the generation node.

[0032] Subsequently, the data is transmitted to the processing node, i.e. the customer service processing module, and the transmission time is the time point from the generation node sending data to the processing node receiving data. In the processing node, the customer service personnel processes the feedback content, and at this time the processing state is in processing.

[0033] After processing is completed, the data is transmitted to the storage node, i.e. the database storage module, and the transmission time is the time point from the processing node sending data to the storage node receiving data, and the processing state becomes completed.

[0034] In addition, after the transaction record dataset is generated from the generation node (transaction system module), it is first transmitted to the processing node (transaction analysis module) for data analysis, and the processing state is in processing, and then it is transmitted to the storage node (transaction database) for storage, and the processing state becomes completed, and the transmission time of each node has a corresponding record.

[0035] Extract these node identifiers, transmission times and processing states, arrange them in chronological order, and generate the flow feature of each data asset.

[0036] Step S115: Perform deduplication verification processing on the structure features, the association features and the flow features, eliminate the feature information of duplicate records, associate the verified feature information according to the unique code of the data asset, and generate the original feature information.

[0037] After the structural features, the correlation features and the flow features are collected, a de-duplication check needs to be performed. For the structural features, it is checked whether there are duplicate field level relationship descriptions, for example, whether there are two identical tree structure records in the structural features of the customer basic information dataset. If there are, the duplicate records are removed.

[0038] The de-duplication check of the correlation features mainly checks whether there are duplicate correlation relationship records, for example, whether the reference relationship between the customer basic information dataset and the transaction record dataset is recorded multiple times. If there are duplicates, only one record is retained.

[0039] The de-duplication check of the flow features is directed to the duplicate records of the same data asset in the same node and the same processing state, for example, whether there are duplicate delivery records in the completed state of the customer feedback dataset in the storage node. If there are duplicates, they are removed.

[0040] After the de-duplication check is completed, each data asset has a unique code, for example, the code of the customer basic information dataset is DS001, and the code of the transaction record dataset is DS002. The structural features, the correlation features and the flow features after the check are associated according to these unique codes, so that each code corresponds to a complete set of feature information, thereby generating the original feature information.

[0041] Step S120: performing a visual dimension mapping process on the original feature information to obtain a visual dimension set of the data asset, the visual dimension set including a structure dimension, a correlation dimension and a flow dimension.

[0042] After obtaining the original feature information, it needs to be mapped to the visual dimensions in order to subsequently construct a display template. This process needs to be performed for the structural features, the correlation features and the flow features respectively to ensure that the mapped dimensions can accurately reflect the features of the data asset.

[0043] Step S121: analyzing the structural features in the original feature information, identifying the level depth, the number of nodes and the connection strength between the nodes in the hierarchical tree form, mapping the level depth to the level axis of the structure dimension, the number of nodes to the scale axis of the structure dimension, and the connection strength between the nodes to the correlation axis of the structure dimension, to form the structure dimension.

[0044] The structural features of the customer basic information dataset are analyzed, and the level depth of the hierarchical tree form is 3 levels, the root node is the first level, the personal information and the enterprise information sub-nodes are the second level, and the field leaf nodes are the third level. The level depth is mapped to the level axis of the structure dimension, and the scale of the level axis corresponds to the numerical value of the level depth. The level of the structure can be intuitively reflected through the scale on the axis.

[0045] In terms of the number of nodes, in the hierarchical tree structure of the customer basic information dataset, there is 1 root node, 2 nodes in the second layer, and 8 nodes in the third layer, and the total number of nodes is 11. The number of nodes is mapped to the scale axis of the structure dimension, and the value range of the scale axis corresponds to the number of nodes. The more the number of nodes, the more to the right on the scale axis.

[0046] The connection strength between nodes is determined by the dependency relationship and the association frequency between fields. For example, the connection strength between the customer name field and the contact information field is strong because there is a direct dependency relationship and they are often queried at the same time. The connection strength between the gender field and the enterprise size field is weak, and there is almost no direct association between them. These connection strengths are mapped to the association axis of the structure dimension, and the numerical size of the association axis corresponds to the strength of the connection strength.

[0047] The hierarchical axis, the scale axis and the association axis together constitute the structure dimension of the customer basic information dataset. The structure dimensions of other data assets are also generated in the same way.

[0048] Step S122: analyze the associated features in the original feature information, extract the frequency of reference relationship, the coverage of sharing relationship and the complexity of combination relationship, the frequency of reference relationship refers to the number of references in unit time, the coverage of sharing relationship refers to the proportion of shared fields in total fields, and the complexity of combination relationship refers to the number of data assets required for combination, and is respectively mapped to the frequency axis, the coverage axis and the complexity axis of the association dimension, forming the association dimension.

[0049] The reference relationship between the customer basic information dataset and the transaction record dataset is analyzed, and the number of references of the customer identification field in the transaction record dataset to the customer identification field in the customer basic information dataset in unit time is counted to obtain the frequency of the reference relationship. The frequency is mapped to the frequency axis of the association dimension, and the higher the frequency, the more to the front on the frequency axis.

[0050] For the sharing relationship, taking the customer identification field as an example, the proportion in the total number of fields of the customer basic information dataset is a certain percentage, the proportion in the total number of fields of the transaction record dataset is another percentage, and the proportion in the total number of fields of the customer feedback dataset is a percentage. Take the average of these proportions as the coverage of the sharing relationship. The coverage is mapped to the coverage axis of the association dimension, and the higher the coverage, the more upward on the coverage axis.

[0051] In terms of the complexity of the combination relationship, the customer value analysis dataset is composed of 3 data assets, so the complexity of its combination relationship is 3. The complexity is mapped to the complexity axis of the association dimension, and the numerical value of the complexity axis corresponds to the number of data assets required for combination.

[0052] The correlation dimension is composed of the frequency axis, the coverage axis and the complexity axis, and is used to describe the correlation characteristics between the data assets.

[0053] Step S123: Processing the flow characteristics in the original feature information, extracting the conversion sequence of node identification, the interval distribution of transmission time and the conversion probability of processing state, the conversion sequence of node identification refers to the order of nodes passed by the data asset, the interval distribution of transmission time refers to the distribution of time difference between adjacent nodes, and the conversion probability of processing state refers to the possibility of conversion from one state to another, and is mapped to the node axis, the time axis and the state axis of the flow dimension, forming the flow dimension.

[0054] The flow characteristics of the customer feedback data set are processed, and the conversion sequence of node identification is generation node-processing node-storage node. The conversion sequence is mapped to the node axis of the flow dimension, and each node is marked in order on the node axis to show the flow path of the data asset.

[0055] The interval distribution of transmission time refers to the distribution of time difference of the customer feedback data set from the generation node to the processing node, and the time difference from the processing node to the storage node. For example, most of the time differences are concentrated in a certain range, and a few time differences are outside the range. The above distribution is mapped to the time axis of the flow dimension, and the distribution of time difference is reflected by different interval division on the time axis.

[0056] In terms of conversion probability of processing state, the probability of conversion of the customer feedback data set from the to-be-processed state to the processing state is high, the probability of conversion from the processing state to the completed state is also high, and the probability of conversion from the completed state to other states is extremely low. These conversion probabilities are mapped to the state axis of the flow dimension, and the distance between states on the state axis corresponds to the size of the conversion probability. The closer the distance, the higher the conversion probability.

[0057] The flow dimension is composed of the node axis, the time axis and the state axis, and is used to reflect the flow characteristics of the data asset.

[0058] Step S124: Dimension calibration processing is performed on the structure dimension, the correlation dimension and the flow dimension, and the calibrated structure dimension, the correlation dimension and the flow dimension are integrated into a set of visualized dimensions.

[0059] The dimension calibration processing mainly ensures that the structure dimension, the correlation dimension and the flow dimension are consistent and coordinated when displayed. For the hierarchy axis, the size axis and the correlation axis of the structure dimension, the scale range of each axis is adjusted, so that the structure dimensions of different data assets can be compared in the same range. For example, the hierarchy axis scale of all data assets is uniformly set to 0 to 5, so that the customer basic information data set with a hierarchy depth of 3 and the transaction record data set with a hierarchy depth of 2 can be displayed on the same axis.

[0060] The frequency axis, coverage axis and complexity axis of the association dimension also need to be calibrated to unify the measurement standards of each axis. For example, the unit of the frequency axis is unified as times / hour, the value range of the coverage axis is unified as 0 to 1, and the numerical range of the complexity axis is adjusted according to the maximum number of the actual combined data assets.

[0061] The calibration of the node axis, time axis and state axis of the flow dimension mainly unifies the format of the node label, the time unit of the time axis (such as unified as minutes), and the conversion probability representation of the state axis.

[0062] After the calibration process, the structure dimension, the association dimension and the flow dimension are integrated together to form a visual dimension set.

[0063] Step S130: According to the visual dimension set, a multi-level display template of the data asset is constructed, and the multi-level display template includes a basic display layer, an association display layer and a dynamic display layer.

[0064] Based on the visual dimension set, the multi-level display template is constructed, which needs to fully utilize the information of the structure dimension, the association dimension and the flow dimension to construct the basic display layer, the association display layer and the dynamic display layer respectively, and set the corresponding level switching rules, so that each display layer can be organically combined to provide comprehensive visual display for the user.

[0065] Step S131: Based on the structure dimension in the visual dimension set, the basic display layer is constructed, and the basic display layer adopts a tree diagram form to display the structural characteristics of the data asset. The root node of the tree diagram corresponds to the highest level of the data asset, the child nodes correspond to the next level, the size of the node is determined according to the scale axis of the structure dimension, and the thickness of the connection line between the nodes is determined according to the association axis of the structure dimension.

[0066] The tree diagram of the basic display layer is constructed based on the structure dimension of the customer basic information data set. The root node of the tree diagram is the customer basic information data set, corresponding to the highest level. The child nodes under the root node are personal information and enterprise information, corresponding to the next level. The child nodes under the personal information child node are name, gender, age, contact information, etc. The child nodes under the enterprise information child node are enterprise name, industry, enterprise size, etc.

[0067] The size of the node is determined according to the scale axis of the structure dimension. The larger the scale axis value is, the larger the node is. The total number of nodes of the customer basic information data set is 11, and there is a corresponding value on the scale axis. The size of the root node is determined according to the value, which is larger than the size of the child node, because the number of nodes corresponding to the child node is smaller.

[0068] The thickness of the connection between nodes is determined according to the correlation axis of the structure dimension, and the greater the value of the correlation axis, the thicker the connection. The connection strength between the customer name field and the contact information field has a greater value on the correlation axis, so the connection between the two is thicker; the connection strength between the gender field and the enterprise size field has a smaller value, so the connection is thinner.

[0069] Through the above tree diagram form, the structural characteristics of the customer basic information dataset are clearly displayed on the basic display layer, and the structural characteristics of other data assets are also displayed on the basic display layer in the same way.

[0070] Step S132: According to the correlation dimension in the set of visualization dimensions, a correlation display layer is constructed on the basis of the basic display layer, the correlation display layer is superimposed on the tree diagram in the form of a network diagram, the nodes in the network diagram correspond to data assets, the connection between nodes represents the correlation relationship, the length of the connection is determined according to the frequency axis of the correlation dimension, the color of the connection is determined according to the coverage axis of the correlation dimension, and the shape of the node is determined according to the complexity axis of the correlation dimension.

[0071] On the basis of the tree diagram of the basic display layer, the correlation display layer in the form of a network diagram is superimposed. The nodes in the network diagram correspond to the customer basic information dataset, the transaction record dataset, the customer feedback dataset, and the customer value analysis dataset, etc.

[0072] The connection between nodes represents the correlation relationship, and the length of the connection between the customer basic information dataset and the transaction record dataset is determined according to the frequency axis of the correlation dimension. Since the reference relationship between the two is high in frequency, the value of the frequency axis is large, so the connection is short.

[0073] The color of the connection is determined according to the coverage axis of the correlation dimension. The customer basic information dataset, the transaction record dataset, and the customer feedback dataset share the customer identification field, and the coverage axis value is high, so the color of the connection between the three is dark; the coverage axis value of other correlation relationships is low, and the connection color is light.

[0074] The shape of the node is determined according to the complexity axis of the correlation dimension. The complexity of the combination relationship of the customer value analysis dataset is 3, and the complexity axis value is large, so the shape of the node is hexagonal; the complexity axis value of the customer basic information dataset and other nodes with small values is small, so the shape is circular.

[0075] Through the above network diagram form of the correlation display layer, the correlation relationship between each data asset is clearly displayed on the basis of the basic display layer.

[0076] Step S133: in combination with the flow dimension in the set of visualization dimensions, a dynamic display layer is constructed, which displays the flow characteristics of the data asset in the form of a timeline animation, the scale of the timeline is determined according to the timeline of the flow dimension, the icon in the timeline animation represents the data asset, the moving path of the icon is determined according to the node axis of the flow dimension, and the color change of the icon is determined according to the state axis of the flow dimension.

[0077] In combination with the flow dimension, the dynamic display layer is constructed, taking the flow of the customer feedback dataset as an example. The scale of the timeline is determined according to the timeline of the flow dimension, and each scale on the timeline represents a certain time interval (such as 10 minutes).

[0078] In the timeline animation, a specific icon represents the customer feedback dataset, and the moving path of the icon is determined according to the node axis of the flow dimension, i.e. moving from the generation node icon to the processing node icon, and then to the storage node icon, clearly showing the flow path.

[0079] The color change of the icon is determined according to the state axis of the flow dimension, when in the processing state, the icon is gray; when entering the processing state, the icon becomes blue; when completing the processing and entering the completed state, the icon becomes green. In the state conversion process, the color will be smoothly transitioned to reflect the conversion of the state.

[0080] For other data assets such as transaction record datasets, their flow characteristics are also displayed in the form of a timeline animation in the dynamic display layer in the same way. For example, the icon of the transaction record dataset moves from the transaction system module icon (generation node) to the transaction analysis module icon (processing node), and then to the transaction database icon (storage node), and the color also changes from gray in the processing state to blue in the processing state, and finally to green in the completed state.

[0081] Step S134: set the level switching rules of the multi-level display template, the basic display layer is the default display layer, and through the preset switching instruction, it can be switched to the associated display layer or the dynamic display layer, and the switching process adopts a smooth transition animation, and the basic display layer, the associated display layer, the dynamic display layer and the level switching rules are integrated into a multi-level display template.

[0082] The level switching rules of the multi-level display template need to be clear and easy to operate. The basic display layer is set as the default display layer, when the user opens the visualization interface, the tree chart of the basic display layer is first presented.

[0083] The preset switching instruction includes two forms of shortcut keys and interface buttons. For example, the user presses the "G" key on the keyboard or clicks the "correlation view" button at the top of the interface to switch from the basic display layer to the correlation display layer; presses the "D" key or clicks the "dynamic view" button to switch to the dynamic display layer; and presses the "B" key or clicks the "basic view" button to switch back to the basic display layer from other display layers.

[0084] The switching process adopts a smooth transition animation. When switching from the basic display layer to the correlation display layer, the tree diagram gradually fades out, and the network diagram gradually becomes clear, and the entire transition process lasts for a certain time (such as 0.5 seconds), so as to avoid interface mutation and bring discomfort to the user. When switching from the correlation display layer to the dynamic display layer, the network diagram gradually fades out, and the timeline animation gradually appears; when switching from the dynamic display layer to other display layers, the animation gradually stops and fades out, and the content of the target display layer gradually becomes clear.

[0085] The basic display layer, the correlation display layer, the dynamic display layer, and the above-mentioned layer switching rules are integrated together to form a complete multi-level display template.

[0086] Step S140: According to the analysis requirements of the user, the multi-level display template is interactively adapted to generate a visual interactive interface of the data asset.

[0087] After obtaining the multi-level display template, it is necessary to interactively adapt according to the analysis requirements of the user, so that the display content is more suitable for the use scene of the user, and the efficiency of the user in analyzing the data asset is improved. This process involves analysis of user requirements, selection of display layers, processing of focus objects, adjustment of display precision, and addition of interactive controls, and many other aspects.

[0088] Step S141: Receive the analysis requirement instruction input by the user, analyze the analysis type, focus object, and display precision requirement contained in the analysis requirement instruction, the analysis type includes structure analysis, correlation analysis, and flow analysis, the focus object refers to the data asset that the user focuses on, and the display precision requirement refers to the detail level of the display content.

[0089] The system receives the analysis requirement instruction of the user through the input box, the drop-down menu or the voice input module on the interface. For example, the user inputs "analyze the correlation between customer basic information and transaction records, focus on customer A, and display detailed information" in the input box.

[0090] It can be known from the analysis of the instruction that the analysis type is correlation analysis; the focus object is the data asset corresponding to customer A, that is, the basic information data set of customer A and the related transaction record data set; and the display precision requirement is the detailed level, which requires to display more fields and correlation details.

[0091] If the user inputs "view the flow of the customer feedback dataset and show brief information", the analysis type is flow analysis, the focus object is the customer feedback dataset, and the display accuracy requirement is briefness, which only needs to show the key flow nodes and states.

[0092] Step S142: According to the analysis type, determine the corresponding display layer in the multi-level display template. If it is structure analysis, take the basic display layer as the main one; if it is association analysis, take the association display layer as the main one; if it is flow analysis, take the dynamic display layer as the main one.

[0093] When the analysis type is structure analysis, the system determines to take the basic display layer as the main one for display. For example, the user needs to analyze the field hierarchical structure of the customer basic information dataset, at which time the interface mainly presents the tree diagram of the basic display layer, clearly showing the hierarchical nodes and field relationships.

[0094] If the analysis type is association analysis, such as the user needs to analyze the reference relationship between the customer basic information dataset and the transaction record dataset, take the association display layer as the main one, and the interface mainly presents the network diagram superimposed on the tree diagram, highlighting the connection and association attributes between the two.

[0095] When the analysis type is flow analysis, such as analyzing the delivery process of the customer feedback dataset, the system takes the dynamic display layer as the main one, and the interface mainly presents the time axis animation to show the flow of the data asset at each node.

[0096] Step S143: For the focus object, perform highlight processing in the corresponding display layer, including magnification display, color change, and adding a border.

[0097] For the focus object being the customer A's basic information dataset and the related transaction record dataset, perform highlight processing in the association display layer. Magnify the customer A's basic information dataset node and the related transaction record dataset node by a certain percentage larger than other nodes; at the same time, change the node color to bright orange, which is distinguished from the color of other nodes; and add a red border around the node to further highlight it.

[0098] If the focus object is the customer feedback dataset, in the dynamic display layer, the icon representing the dataset is magnified, the color is changed to yellow, and a blue border is added, so that the user can quickly identify it.

[0099] Step S144: According to the display accuracy requirement, adjust the detail display degree of the display layer.

[0100] When the display accuracy requirement is detailed, in the basic display layer, each node of the tree diagram not only displays the field name, but also displays the format type, length, and other additional information of the field; in addition to the thickness indicating the connection strength, the connection between nodes also marks the specific dependency relationship description.

[0101] In the association display layer, the connection of the network graph is annotated with specific values of frequency, names of shared fields, brief descriptions of combination rules, etc. In addition to shape differentiation, the node also displays the unique code of the data asset and the number of main fields.

[0102] If the display precision requirement is brief, the tree graph of the basic display layer only displays the main hierarchical nodes and key field names; the network graph of the association display layer only retains important association connections and does not display specific values and descriptions; the timeline animation of the dynamic display layer only displays the main flow nodes, simplifying the details of state transitions.

[0103] Step S145: Add interactive controls to the visualization interactive interface, including zoom controls, filtering controls, detail controls, and comparison controls. Zoom controls are used to adjust the size of the display content, filtering controls are used to filter data assets with specific conditions, detail controls are used to view detailed information of data assets, and comparison controls are used to compare features of multiple data assets.

[0104] Step S1451: Add zoom controls to the edge area of the visualization interactive interface, including zoom-in buttons, zoom-out buttons, and zoom sliders. Zoom-in buttons display the content at a preset scale, zoom-out buttons display the content at a preset scale, and zoom sliders continuously adjust the zoom scale of the display content by dragging the position. The zoom range is limited between the preset minimum zoom scale and the maximum zoom scale.

[0105] Add zoom controls to the right edge area of the interface. The zoom-in button is an upward arrow icon, the zoom-out button is a downward arrow icon, and the zoom slider is a rectangular block that can be dragged up and down.

[0106] When the user clicks the zoom-in button, the display content is zoomed in by a preset scale (e.g. 20% each time), making the details clearer. Clicking the zoom-out button zooms out by the same scale, displaying more overall content.

[0107] The drag range of the zoom slider corresponds to the range of the zoom scale. The preset minimum zoom scale is 50% and the maximum zoom scale is 200%. When the slider is at the lower end, the display content is at the minimum zoom scale; when it is dragged to the upper end, it is at the maximum zoom scale, and the intermediate position corresponds to the corresponding intermediate scale.

[0108] Step S1452: Add a filter control in the top area of the visualization interface, which includes a structure filter option, an association filter option, and a flow filter option. The structure filter option allows users to select a specific structure level or format type. The association filter option allows users to set a frequency threshold or coverage threshold for association relationships. The flow filter option allows users to specify a flow node or processing state. After selecting the filter option, the layer automatically hides data assets that do not meet the filter conditions.

[0109] In the top area of the interface, set a drop-down menu for the filter control. In the drop-down menu of the structure filter option, include structure level options such as "Level 1", "Level 2", "Level 3", etc., and format type options such as "Structured Data", "Unstructured Data", "Text Format", "Integer Format", etc. After the user checks the corresponding options, the data asset nodes in the display layer that do not meet the selected structure level or format type will be hidden.

[0110] In the association filter option, users can set the frequency threshold of the reference relationship through the input box (e.g., input "5" to display only the association relationships with a frequency higher than 5 times / hour), and set the coverage threshold of the shared relationship through the slider bar (e.g., slide to 80% to display only the shared relationships with a coverage higher than 80%).

[0111] In the drop-down menu of the flow filter option, include flow node options such as "Generation Node", "Processing Node", "Storage Node", etc., and processing state options such as "Pending", "Processing", "Completed", etc. After the user selects, the data assets in the display layer that do not contain the specified nodes or states will be hidden.

[0112] Step S1453: Add a details control in the right area of the visualization interface, which is in the form of a floating button. When the user clicks on a data asset node in the display layer, the details control automatically activates and displays the original feature information of the data asset, including the complete level of structure features, detailed relationships of association features, and full records of flow features. The details control also includes a close button that can hide the details information when clicked.

[0113] The details control is a floating button with the "Details" text, located in the middle of the right side of the interface, and the initial state is gray and not clickable. When the user clicks on the customer basic information dataset node in the display layer, the details control becomes blue and clickable, and an information panel is automatically popped up.

[0114] The original feature information is displayed in the information panel in the order of structural features, associated features, and flow features. The structural feature part displays a complete hierarchical tree diagram, including all fields and dependency relationships; the associated feature part lists the details of the reference, sharing, and combination relationship with other data sets; and the flow feature part displays the whole transfer record in chronological order, including the identification, transfer time, and processing status of each node.

[0115] In the upper right corner of the information panel, there is a "X" shaped close button. When the user clicks it, the panel is hidden, and the detail control is restored to a gray and non-clickable state.

[0116] Step S1454: Add a comparison control in the bottom area of the visualization interactive interface, which contains an add comparison button and a comparison panel. When the user clicks the add comparison button, the currently selected data asset is added to the comparison list, and the comparison panel displays the feature comparison of all data assets in the comparison list. The comparison panel also contains a clear button for clearing the comparison list.

[0117] In the bottom area of the interface, the add comparison button is a button with the text "+ comparison", and the comparison panel is an expandable rectangular area.

[0118] After the user selects the customer basic information data set node and clicks the add comparison button, the data set is added to the comparison list. Then, after selecting the transaction record data set node and clicking the add comparison button, the transaction record data set is also added to the comparison list.

[0119] At this time, the comparison panel is expanded, and the comparison content is displayed in a table form in the panel. The columns of the table are data asset name, structure dimension (hierarchical axis, size axis, and association axis), association dimension (frequency axis, coverage axis, and complexity axis), and flow dimension (node axis, time axis, and state axis). Each row corresponds to the index values of each data asset.

[0120] In the lower right corner of the comparison panel, there is a "clear" button. When the user clicks it, all data assets in the comparison list are removed, and the table content is cleared.

[0121] Step S146: Set the response rules of the interactive control. When the user operates the interactive control, the corresponding display layer is updated according to the preset rules. The filter control triggers the hiding of data assets that do not meet the conditions. The adapted display layer, interactive control, and response rules are integrated into the visualization interactive interface.

[0122] Step S1461: For the zoom control, set the scaling adjustment rule. When the user clicks the zoom-in button, the zoom ratio of the displayed content increases by a preset increment on the current basis, and the increment is the same each time. When the user clicks the zoom-out button, the zoom ratio of the displayed content decreases by a preset increment on the current basis until it reaches the minimum zoom ratio and is no longer responsive. When the user drags the zoom slider, the zoom ratio changes linearly with the slider position. The zoom slider is at the leftmost end when the minimum zoom ratio is reached, and the rightmost end when the maximum zoom ratio is reached. The center position of the displayed content remains unchanged during the zooming process.

[0123] The preset increment is 20%, the maximum zoom ratio is 200%, and the minimum zoom ratio is 50%. When the current zoom ratio is 100%, the user clicks the zoom-in button, and the ratio becomes 120%. Clicking again changes it to 140%, and so on until it reaches 200%. Continuing to click the zoom-in button will no longer change it.

[0124] When the zoom-out button is clicked, the current ratio of 100% changes to 80%, and then to 60% when clicked again. It will no longer respond after reaching 50%.

[0125] When the zoom slider is dragged, the slider position and the zoom ratio are linearly related. During the process of dragging the slider from the leftmost end (50%) to the rightmost end (200%), the zoom ratio changes uniformly. During the entire zooming process, the center of the displayed content remains in the center of the interface, avoiding content offset.

[0126] Step S1462: For the filtering control, set the hiding rule. When the user selects the structure filtering option, the data asset nodes and their connections in the display layer that do not belong to the selected structure level or format type are automatically hidden. The hiding process uses a fade-out animation, and the remaining nodes automatically adjust their positions. When the user selects the association filtering option, the connections in the display layer whose frequency or coverage is lower than the corresponding preset threshold are automatically hidden, and the data asset nodes are retained but their colors change to light tones. When the user selects the flow filtering option, the data asset icons and their movement paths in the display layer that do not contain the specified flow nodes or processing states are automatically hidden, and the time axis range is adjusted according to the flow time of the remaining data assets.

[0127] After the user selects "Level 2" and "Structured Data" in the structure filtering option, the data asset nodes and their connections in the display layer that belong to Level 1, Level 3, and unstructured data begin to fade out. The animation lasts for 0.3 seconds before completely hiding. The remaining nodes automatically adjust their positions to make the interface layout more reasonable and avoid empty areas.

[0128] When the user sets the frequency threshold to 5 times / hour in the association filtering option, all reference relationship lines with a frequency below 5 times / hour in the association display layer gradually disappear, and the corresponding node color changes to light gray, which contrasts with the node color that meets the conditions.

[0129] When the user selects the "in processing" state in the flow filtering option, the data asset icons in the dynamic display layer in the to-be-processed and completed states and their movement paths are hidden, and the start and end times of the timeline are adjusted according to the flow time of the data assets in the processing state, so that the display content is more focused.

[0130] Step S1463: Regarding the detail control, a display rule is set, when the user clicks the data asset node, the detail control slides out from the right side of the interface, the sliding-out animation lasts for a preset time, and the detail information is loaded and displayed in the order of structure characteristics, association characteristics, and flow characteristics; when the user clicks the close button, the detail control slides into the right side to hide, the sliding-in animation lasts for a preset time, and the selected state of the data asset node is released.

[0131] The duration of the preset sliding-out and sliding-in animation is 0.4 seconds. When the user clicks the customer feedback dataset node, the detail control slides out from the right edge of the interface, gradually displaying the complete information panel. The information in the panel is loaded in the order of the hierarchical tree diagram of the structure characteristics, the relationship list of the association characteristics, and the transfer record of the flow characteristics, and the loading interval of each part is 0.1 seconds.

[0132] After the user clicks the close button, the information panel slides into the right side, gradually hiding in the edge of the interface, and the animation lasts for 0.4 seconds. At the same time, the selected state of the customer feedback dataset node is released, the highlight effect of the node disappears, and the node returns to the normal display state.

[0133] Step S1464: For the comparison control, a list update rule is set, when the user clicks the add comparison button, if the data asset is not in the comparison list, it is added to the comparison list and a corresponding row is added to the comparison panel; if it is already in the comparison list, a prompt information is popped up to indicate that the data asset is already in the comparison list; when the user clicks the clear button, the comparison list is emptied, the table content in the comparison panel is also emptied, and the selected state of all data asset nodes is released.

[0134] The user selects the customer basic information dataset and clicks the add comparison button. Since the dataset is not in the comparison list, the system adds it to the list, and a row of index values of the customer basic information dataset is added to the table in the comparison panel.

[0135] If the user selects the customer basic information dataset again and clicks the add comparison button, the system will pop up a prompt box, prompting "the data asset has been in the comparison list", and the prompt box will disappear automatically after 3 seconds.

[0136] When the user clicks the clear button, all data assets in the comparison list are removed, the contents in the table are emptied, and the highlight effect of all selected data asset nodes disappears, returning to the normal state.

[0137] The display layer after interactive adaptation processing, the added interactive controls, and the set response rules are integrated to form a visual interactive interface of data assets, facilitating the user's analysis operation of data assets.

[0138] Step S150: Track the user's operation trajectory data in the visual interactive interface, adjust the multi-level display template based on the operation trajectory data, and update the visual interactive interface.

[0139] In order to make the visual interactive interface more in line with the user's usage habits, it is necessary to track the user's operation trajectory, analyze the user's behavior preferences, and then adjust the multi-level display template, update the interface content, and improve the user experience.

[0140] Step S151: Embed a trajectory collection module in the visual interactive interface, which records all user operations in real time, including mouse clicks, mouse movements, keyboard inputs, and control operations.

[0141] The trajectory collection module is embedded in the visual interactive interface in the form of a code plug-in, and the module runs in the background after starting, without affecting the user's normal operation. This trajectory collection module can capture all user operations on the interface, including mouse clicks on any position or element on the interface, mouse movement trajectories on the interface, any characters input through the keyboard, and operations on various interactive controls.

[0142] The module generates a record for each operation, including the type of operation, the time of occurrence, the object involved, and other information, and stores it in a local log file in chronological order.

[0143] Step S152: When the user performs a mouse click operation, record the position coordinates of the click, the object identifier of the click, and the click time. The object identifier refers to the unique identifier of the interactive control or data asset node that is clicked.

[0144] When the user clicks the customer basic information dataset node in the association display layer with the mouse, the trajectory acquisition module records the position coordinates (X-axis and Y-axis coordinates with the upper left corner of the interface as the origin) of the click operation, the object identifier is the unique code of the customer basic information dataset (such as DS001), and the click time is the system time accurate to milliseconds.

[0145] When the user clicks the zoom-in button, the recorded position coordinates are the coordinates of the zoom-in button on the interface, the object identifier is "zoom-in button", and the click time is recorded at the same time.

[0146] Step S153: When the user performs a mouse moving operation, the trajectory acquisition module records the moving path coordinate sequence of the mouse pointer in the interface and the stop point coordinates and stop time in the moving process. The stop point refers to the position where the mouse pointer moving speed is lower than the preset threshold.

[0147] The preset mouse moving speed threshold is a certain pixel / millisecond. When the user moves the mouse from the zoom control to the customer feedback dataset node, the trajectory acquisition module records a series of coordinates passed by the mouse pointer, forming a moving path coordinate sequence.

[0148] During the moving process, if the moving speed of the mouse pointer in the vicinity of the customer basic information dataset node is lower than the preset threshold, the position is determined as a stop point. The trajectory acquisition module will record the X-axis and Y-axis coordinates of the stop point, and start timing from when the mouse pointer arrives at the point until it leaves the point, and record the stop time. For example, when the user views the customer basic information dataset node, the mouse pointer stays above the node for 3 seconds, and the module accurately records the 3-second duration and the corresponding coordinate position.

[0149] Step S154: When the user performs a keyboard input operation, record the input character content, input start time and input end time, and the character content includes the filtering condition or search keyword input in the filtering control.

[0150] When the user inputs "transaction amount greater than a specific value" as the filtering condition in the input box of the filtering control, the trajectory acquisition module records the complete character content input. At the same time, the time when the user presses the first character is recorded as the input start time, and the time when the user presses the enter key or clicks the area outside the input box to confirm the input is recorded as the input end time. If the user inputs "customer B" in the search box for search, the module also records the character content and the corresponding input start and end time.

[0151] Step S155: When the user performs a control operation, record the control type, operation parameters and interface state changes before and after the operation, and the operation parameters include the zoom ratio change value of the zoom control and the filtering condition setting value of the filtering control.

[0152] When the user operates the zoom control to enlarge the button, the control type is recorded as "zoom control-enlarging button", and the operation parameter is the change value of the current zoom ratio (e.g. from 100% to 120%, the change value is 20%). At the same time, the display state of the interface before operation (e.g. the size and position of each node) and the display state of the interface after operation are recorded.

[0153] When the user sets the coverage threshold of the sharing relationship in the filter control to 80%, the control type is recorded as "filter control-associated filter option", the operation parameter is "coverage threshold = 80%", and the display state changes of the data asset nodes and the connection lines in the display layer before and after operation are recorded.

[0154] Step S156: The above recorded operation behaviors are integrated in chronological order to generate operation trajectory data containing operation type, operation object, operation time and operation parameter. The operation trajectory data is stored in the form of a log, and each log entry contains a unique trajectory number.

[0155] The trajectory collection module arranges the records of mouse clicks, mouse movements, keyboard inputs and control operations in chronological order, and each record contains operation type (such as "mouse click", "keyboard input", etc.), operation object (such as "customer basic information data set node", "enlarging button", etc.), operation time (accurate to milliseconds) and operation parameter (such as coordinates, character content, ratio change value, etc.).

[0156] The generated operation trajectory data is stored in the form of a log file in the directory specified by the system, and each log entry has a unique trajectory number, which is composed of date, time and random sequence to ensure the uniqueness of each record. For example, a log entry may be "trajectory number: 20240520153022001 | operation type: mouse click | operation object: customer feedback data set node | operation time: 2024-05-20 15:30:22.123 | operation parameter: X=300, Y=200".

[0157] Step S157: Analyze the mouse click records in the operation trajectory data, and count the click frequencies of different interactive controls and data asset nodes. The objects with click frequencies higher than the preset click threshold are determined as high-frequency operation objects.

[0158] For example, step S1571: filter all mouse click records from the operation trajectory data, and each mouse click record contains the object identifier of the click, the click time and the position coordinates of the click.

[0159] From the stored operation trajectory log, all records of the operation type "mouse click" are filtered out. Each of these records clearly contains the object identifier (such as "DS001", "zoom-in button", etc.) clicked, the specific time when the click occurred, and the position coordinates at the time of the click.

[0160] For example, the filtered records can include "object identifier: DS001, click time: 2024-05-20 15:30:22.123, coordinates: X=300, Y=200" "object identifier: zoom-in button, click time: 2024-05-20 15:31:05.456, coordinates: X=500, Y=100", etc.

[0161] Step S1572: Group the mouse click records according to the object identifier, and group all click records of the same object identifier into a group.

[0162] According to the different object identifiers, the filtered mouse click records are grouped. All click records with the object identifier "DS001" (customer basic information dataset node) are grouped into one group; all click records with the object identifier "zoom-in button" are grouped into another group; and so on, to ensure that the click records corresponding to each object identifier are concentrated in their respective groups.

[0163] Step S1573: For each group of click records corresponding to an object identifier, count the total number of records, calculate the ratio of the total number of clicks to the length of time covered by the records, and obtain the click frequency of the object. The length of time is the difference between the earliest click time and the latest click time in the click record group.

[0164] Taking the click record group with the object identifier "DS001" as an example, the total number of records in this click record group is 20. The earliest click time in this click record group is 2024-05-20 15:30:22.123, and the latest click time is 2024-05-20 16:30:22.123, with a time length of 1 hour (i.e. 3600 seconds). The click frequency of this object is 20 times / 3600 seconds, and the click frequency per second is calculated.

[0165] Similarly, for the click record group of the "zoom-in button", if the total number is 15 and the time length is 2 hours (7200 seconds), the click frequency is 15 times / 7200 seconds.

[0166] Step S1574: Set a preset click threshold, and the preset click threshold is the average value of the click frequencies of all objects.

[0167] An average value of the click frequency of all objects (including each data asset node and interactive control) is calculated, and the average value is set as a preset click threshold. For example, the total click frequency of all objects is 100 times per second, and the total number of objects is 20, so the preset click threshold is 5 times per second.

[0168] Step S1575: Compare the click frequency of each object with the preset click threshold. If the click frequency is higher than the preset click threshold, mark the object as a candidate high-frequency operation object.

[0169] The click frequency of each object is compared with the preset click threshold (5 times per second) one by one. If the click frequency of "DS001" is 6 times per second, which is higher than the threshold, mark it as a candidate high-frequency operation object. If the click frequency of the "zoom-in button" is 3 times per second, which is lower than the threshold, do not mark it.

[0170] Step S1576: Perform time distribution analysis on the click record of the candidate high-frequency operation object. If the click frequency in the recent preset time period accounts for more than a preset time proportion of the total click frequency, determine it as a high-frequency operation object.

[0171] The preset time period is 1 hour, and the preset time proportion is 60%. For the candidate high-frequency operation object "DS001", analyze its click record. The click frequency in the last 1 hour is 12 times, and the total click frequency is 20 times. 12 / 20=60%, reaching the preset time proportion, so "DS001" is determined as a high-frequency operation object.

[0172] If another candidate high-frequency operation object has a click frequency proportion of 50% in the last 1 hour, which is lower than 60%, it is not determined as a high-frequency operation object.

[0173] Step S1577: Establish an identification file for the high-frequency operation object, which contains object identification, click frequency, recent click time, and position information in the interface.

[0174] An identification file is established for the determined high-frequency operation object "DS001". The file records object identification as "DS001", click frequency as 6 times per second, recent click time as 2024-05-20 16:30:22.123, and position coordinates in the interface as X=300, Y=200, etc.

[0175] Step S158: Adjust the display properties of high-frequency operation objects in the multi-level display template. For high-frequency operation interactive controls, increase their display size in the interface and adjust their position to an easy-to-operate area. For high-frequency operation data asset nodes, enhance their highlight effect in the corresponding display layer.

[0176] For the high-frequency operation of the interactive control, such as "filter control - associated filter option", increase its display size in the interface, increase the width and height by a certain percentage, so that it is easier for users to click; at the same time, adjust its position from the top edge of the interface to the middle area of the top, which is in the range of user operation.

[0177] For the high-frequency operation of the data asset node "DS001", in the basic display layer, the associated display layer and the dynamic display layer, the highlight effect is enhanced, in addition to the original magnification, color change and adding border, the flicker effect (flicker once every 1 second) is added, so that it is more prominent in the interface.

[0178] Step S159: analyze the mouse movement record in the operation trajectory data, extract the main path of mouse movement and the distribution of stay points, determine the visual attention area of the user, and the visual attention area refers to the interface area whose total stay time proportion exceeds the preset proportion.

[0179] All mouse movement records are extracted from the operation trajectory data, including the moving path coordinate sequence and the stay point coordinates and stay time. Through the analysis of the moving path, the main path of the user's mouse movement is found out. These paths are the routes that the user frequently moves in the interface, for example, from the zoom control on the left to the customer basic information dataset node in the middle, and then to the detail control on the right.

[0180] The stay time of all stay points is counted, and the total stay time in each area is calculated. The preset proportion is 40%, and the interface area whose stay time total proportion exceeds 40% is determined as the visual attention area of the user. For example, the stay time total proportion of the area in the middle left of the interface is 45%, so the area is determined as the visual attention area.

[0181] Step S1510: adjust the layout of the multi-level display template, and place the core data asset nodes and important interactive controls in the visual attention area first, and optimize the arrangement of nodes in the display layer.

[0182] Move the core data asset nodes (such as customer basic information dataset node, transaction record dataset node) and important interactive controls (such as filter control, detail control) to the determined visual attention area, so that the user can quickly capture these key elements visually.

[0183] At the same time, optimize the arrangement of nodes in the display layer. For the tree chart of the basic display layer, place the high-frequency operation nodes and core nodes in the upper and middle positions of the tree chart to reduce the mouse movement distance of the user; for the network chart of the associated display layer, adjust the distribution of nodes to make the nodes with close relationship gather together to reduce the intersection of lines.

[0184] Step S1511: Analyze the control operation records in the operation track data, extract the user's commonly used filtering conditions, scaling ratio and comparison combination and save them as default configurations. When the user opens the visual interactive interface again, the default configurations are automatically applied.

[0185] Analyzing the control operation records, it is found that the user often sets the filtering conditions of "shared relationship coverage higher than 80%" and "processing status as processing" in the filtering control; the scaling ratio commonly used in the scaling control is 120%; and the customer basic information dataset and the transaction record dataset are commonly combined for comparison in the comparison control.

[0186] These commonly used filtering conditions, scaling ratio and comparison combination are saved as default configurations. When the user opens the visual interactive interface again, the system automatically applies these default configurations, the display layer displays the data assets according to the commonly used filtering conditions, the scaling ratio is automatically adjusted to 120%, and the customer basic information dataset and the transaction record dataset are automatically added to the comparison list.

[0187] Step S1512: Update the layout, display attributes and default configurations of the visual interactive interface according to the adjusted multi-level display template, and use a smooth transition animation in the updating process to complete the updating of the visual interactive interface.

[0188] According to the adjusted multi-level display template, the visual interactive interface is updated. In terms of layout, the core nodes and important controls are moved to the visual focus area, and the node arrangement is optimized; in terms of display attributes, the size, position and highlight effect of high-frequency operation objects are adjusted; and in terms of default configurations, the user's commonly used filtering conditions, scaling ratio and comparison combination are applied.

[0189] In the updating process, a smooth transition animation is used, the movement, size change and display state change of interface elements are realized through animation, the duration is 0.5 seconds, and the interface mutation is avoided to affect the user experience. After the animation is completed, the visual interactive interface is updated and presented in a state that is more in line with the user's operation habits.

[0190] Figure 2 A schematic diagram of exemplary hardware and software components of a data asset display and analysis system 100 based on data visualization that can implement the idea of the present application is shown. For example, a processor 120 can be used in the data asset display and analysis system 100 based on data visualization and for performing the functions in the present application.

[0191] The data asset display and analysis system 100 based on data visualization can be a general server or a special-purpose server, both of which can be used to implement the data asset display and analysis method based on data visualization of the present application. The present application only shows one server, but for the sake of convenience, the functions described in the present application can be implemented in a distributed manner on multiple similar platforms to balance the processing load.

[0192] For example, the data asset display and analysis system 100 based on data visualization can include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and different forms of storage media 140, such as a disk, a ROM, or a RAM, or any combination thereof. The data asset display and analysis system 100 based on data visualization can also include program instructions stored in a ROM, a RAM, or other types of non-transitory storage media, or any combination thereof, for example. The method of the present application can be implemented according to these program instructions. The data asset display and analysis system 100 based on data visualization also includes an I / O interface 150 between the computer and other input and output devices.

[0193] For the sake of illustration, only one processor is described in the data asset display and analysis system 100 based on data visualization. However, it should be noted that the data asset display and analysis system 100 based on data visualization in the present application can also include multiple processors, so the steps performed by one processor described in the present application can also be jointly performed or separately performed by multiple processors. For example, if the processor of the data asset display and analysis system 100 based on data visualization performs steps A and B, it should be understood that steps A and B can also be performed by two different processors together or separately in one processor. For example, a first processor performs step A and a second processor performs step B, or the first processor and the second processor jointly perform steps A and B.

[0194] In addition, the present application also provides a readable storage medium, in which computer executable instructions are pre-set, and when the processor executes the computer executable instructions, the data asset display and analysis method based on data visualization as described above is implemented.

[0195] It should be noted that, in order to simplify the description of the present application and to help understand one or more embodiments of the present application, in the foregoing description of the embodiments of the present application, various features are sometimes combined into one embodiment, figure or description thereof.

Claims

1. A data asset display and analysis method based on data visualization, characterized in that, The method includes: Collect the original characteristic information of the data assets, which includes the structural characteristics, association characteristics and circulation characteristics of the data assets; The original feature information is subjected to visualization dimension mapping processing to obtain a set of visualization dimensions for the data asset, which includes structural dimensions, relational dimensions, and flow dimensions. Based on the set of visualization dimensions, a multi-level display template for data assets is constructed, which includes a basic display layer, a related display layer, and a dynamic display layer. Based on the user's analytical needs, the multi-level display template is interactively adapted to generate a visual interactive interface for data assets. Track the user's operation trajectory data in the visual interactive interface, adjust the multi-level display template based on the operation trajectory data, and update the visual interactive interface; The original characteristic information of the collected data assets includes: Access the data asset management system and retrieve the metadata records of all data assets stored in the data asset management system. The metadata records include the field composition, format type and storage path of the data assets. The hierarchical relationships between fields, the dependencies between fields, and the compatibility attributes of format types are extracted from the metadata records and integrated to form the structural features of the data asset, which are presented in a hierarchical tree form. Query the association mapping table in the data asset management system, extract the reference relationship, sharing relationship and combination relationship between different data assets, and integrate them into the association characteristics of data assets. The reference relationship refers to the field call relationship of one data asset to another data asset, the sharing relationship refers to the basic field set shared by multiple data assets, and the combination relationship refers to the aggregation form formed by multiple data assets according to the set rules. Acquire the transfer records of data assets between different processing nodes, extract the node identifier, transfer time and processing status during the transfer process, and generate the flow characteristics of data assets. The node identifier includes the generation node, processing node and storage node, and the processing status includes pending, processing and completed. The structural features, association features, and flow features are deduplicated and verified to remove duplicate feature information. The verified feature information is then associated with the unique code of the data asset to generate the original feature information. The visualization dimension mapping process performed on the original feature information to obtain a set of visualization dimensions for the data assets includes: The structural features in the original feature information are analyzed to identify the hierarchical depth, number of nodes and connection strength between nodes in the hierarchical tree form. The hierarchical depth is mapped to the hierarchical axis of the structural dimension, the number of nodes is mapped to the scale axis of the structural dimension, and the connection strength between nodes is mapped to the association axis of the structural dimension, thus forming the structural dimension. Analyze the association features in the original feature information, extract the frequency of reference relationships, the coverage of sharing relationships, and the complexity of combination relationships. The frequency of reference relationships refers to the number of references per unit time, the coverage of sharing relationships refers to the proportion of the shared field in the total fields, and the complexity of combination relationships refers to the amount of data assets required for combination. These are then mapped to the frequency axis, coverage axis, and complexity axis of the association dimension to form the association dimension. The flow features in the original feature information are processed to extract the transformation sequence of node identifiers, the interval distribution of transmission time, and the transformation probability of processing state. The transformation sequence of node identifiers refers to the order of nodes through which the data asset passes, the interval distribution of transmission time refers to the time difference distribution between adjacent nodes, and the transformation probability of processing state refers to the possibility of changing from one state to another. These are then mapped to the node axis, time axis, and state axis of the flow dimension to form the flow dimension. The structural dimension, the relational dimension, and the flow dimension are calibrated, and the calibrated structural dimension, relational dimension, and flow dimension are integrated into a set of visual dimensions. The construction of a multi-level display template for data assets based on the set of visualization dimensions includes: Based on the structural dimensions in the visualization dimension set, a basic display layer is constructed. The basic display layer uses a tree diagram to display the structural features of the data assets. The root node of the tree diagram corresponds to the highest level of the data assets, and the child nodes correspond to the next level. The size of the nodes is determined according to the scale axis of the structural dimension, and the thickness of the lines between the nodes is determined according to the correlation axis of the structural dimension. Based on the associated dimensions in the visualization dimension set, an associated display layer is constructed on the basis of the basic display layer. The associated display layer adopts a network diagram form superimposed on the tree diagram. The nodes in the network diagram correspond to data assets, the lines between nodes represent the association relationship, the length of the line is determined according to the frequency axis of the associated dimension, the color of the line is determined according to the coverage axis of the associated dimension, and the shape of the node is determined according to the complexity axis of the associated dimension. Combining the flow dimension in the visualization dimension set, a dynamic display layer is constructed. The dynamic display layer uses a timeline animation to display the flow characteristics of data assets. The scale of the timeline is determined according to the timeline of the flow dimension. The icons in the timeline animation represent data assets. The movement path of the icons is determined according to the node axis of the flow dimension. The color change of the icons is determined according to the state axis of the flow dimension. Set the hierarchical switching rules for the multi-level display template. The basic display layer is the default display layer. You can switch to the related display layer or the dynamic display layer through the preset switching command. The switching process uses a smooth transition animation. The basic display layer, related display layer, dynamic display layer and hierarchical switching rules are integrated into a multi-level display template.

2. The data asset display and analysis method based on data visualization according to claim 1, characterized in that, The step of performing interactive adaptation processing on the multi-level display template based on the user's analysis needs to generate a visual interactive interface for data assets includes: The system receives analysis request instructions from users and parses the analysis type, objects of interest, and display precision requirements contained in the analysis request instructions. The analysis type includes structural analysis, correlation analysis, and flow analysis. The objects of interest refer to the data assets that the user is interested in, and the display precision requirements refer to the level of detail of the displayed content. Based on the analysis type, determine the corresponding display layer in the multi-level display template. If it is a structural analysis, the basic display layer is the main focus; if it is a relational analysis, the relational display layer is the main focus; and if it is a flow analysis, the dynamic display layer is the main focus. For the objects of interest, highlight them in the corresponding display layer. Highlighting methods include enlarging the display, changing the color, and adding a border. Adjust the level of detail in the display layer according to the required display precision; Add interactive controls to the visual interactive interface. The interactive controls include zoom controls, filter controls, detail controls, and comparison controls. The zoom controls are used to adjust the size of the displayed content, the filter controls are used to filter data assets based on specific conditions, the detail controls are used to view detailed information about data assets, and the comparison controls are used to compare the characteristics of multiple data assets. Set the response rules for interactive controls. When a user interacts with an interactive control, the corresponding display layer is updated according to preset rules. Data assets that do not meet the conditions for filtering controls are hidden. The adapted display layer, interactive controls, and response rules are integrated into a visual interactive interface.

3. The data asset display and analysis method based on data visualization according to claim 2, characterized in that, The addition of interactive controls to the visual interactive interface includes: A zoom control is added to the edge area of ​​the visual interactive interface. The zoom control includes a zoom-in button, a zoom-out button, and a zoom slider. The zoom-in button triggers the display content to be zoomed in according to a preset ratio, the zoom-out button triggers the display content to be zoomed out according to a preset ratio, and the zoom slider continuously adjusts the zoom ratio of the display content by dragging its position. The zoom range is limited to between the preset minimum zoom ratio and the maximum zoom ratio. Add a filter control to the top area of ​​the visual interactive interface. The filter control includes structure filter options, association filter options and flow filter options. The structure filter option allows users to select a specific structure level or format type. The association filter option allows users to set a frequency threshold or coverage threshold for the association relationship. The flow filter option allows users to specify the flow node or processing status. After selecting a filter option, the display layer automatically hides data assets that do not meet the filter conditions. Add a details control to the right side of the visual interactive interface. The details control is in the form of a floating button. When the user clicks on a data asset node in the display layer, the details control is automatically activated and displays the original feature information of the data asset. The original feature information includes the complete hierarchy of structural features, the detailed relationship of associated features, and the complete record of flow features. The details control also includes a close button, which can be clicked to hide the details information. Add a comparison control to the bottom area of ​​the visual interactive interface. The comparison control includes an add comparison button and a comparison panel. When the user clicks the add comparison button, the currently selected data asset is added to the comparison list. The comparison panel displays the feature comparison of all data assets in the comparison list and presents the indicator values ​​of each data asset in the structural dimension, correlation dimension and circulation dimension in tabular form. The comparison panel also includes a clear button to clear the comparison list.

4. The data asset display and analysis method based on data visualization according to claim 3, characterized in that, The set response rules for interactive controls, when a user interacts with an interactive control, the corresponding display layer updates according to preset rules, and filters controls that trigger the hiding of data assets that do not meet the conditions, including: For the zoom control, set the scaling rules: when the user clicks the zoom button, the zoom level of the displayed content increases by a preset increment, with the same increment for each click, until the maximum zoom level is reached and then no longer responds; when the user clicks the zoom button, the zoom level of the displayed content decreases by a preset increment, until the minimum zoom level is reached and then no longer responds; when the user drags the zoom slider, the zoom level changes linearly with the slider position, with the minimum zoom level at the leftmost end and the maximum zoom level at the rightmost end, and the center position of the displayed content remains unchanged during the zooming process; For the filter control, a hiding rule is set: when the user selects the structure filter option, data asset nodes and their connections that do not belong to the selected structure level or format type in the display layer are automatically hidden, and the hiding process uses a fade-out animation, with the remaining nodes automatically adjusting their positions; when the user selects the association filter option, connections in the display layer whose frequency or coverage of association relationships is lower than the corresponding preset threshold are automatically hidden, while the data asset nodes are retained but their colors change to a lighter shade; when the user selects the flow filter option, data asset icons and their movement paths that do not contain the specified flow node or processing status in the display layer are automatically hidden, and the timeline range is adjusted according to the flow time of the remaining data assets. Regarding the details control, the display rules are set so that when the user clicks on the data asset node, the details control slides out from the right side of the interface, and the slide-out animation lasts for a preset time. The details information is loaded and displayed in the order of structural features, related features, and flow features. When the user clicks the close button, the details control slides into the right side to hide, and the slide-in animation lasts for a preset time. At the same time, the selection of the data asset node is deselected. For the comparison control, set list update rules: when the user clicks the Add Comparison button, if the data asset is not in the comparison list, add it and add a new row of corresponding data in the comparison panel; if it is already in the comparison list, a prompt message will pop up indicating that the data asset is in the comparison list; when the user clicks the Clear button, the comparison list is cleared, the table content in the comparison panel is cleared, and all data asset nodes are deselected.

5. The data asset display and analysis method based on data visualization according to claim 3, characterized in that, The data for tracking the user's operation trajectory in the visual interactive interface includes: A trajectory acquisition module is embedded in the visual interactive interface. The trajectory acquisition module records all user operations in real time, including mouse clicks, mouse movements, keyboard inputs, and control operations. When a user makes a mouse click, the coordinates of the click location, the identifier of the clicked object, and the click time are recorded. The identifier of the clicked object refers to the unique identifier of the interactive control or data asset node that was clicked. When the user moves the mouse, the coordinate sequence of the mouse pointer's movement path in the interface, as well as the coordinates of the dwell points and the dwell time during the movement, are recorded. The dwell point refers to the position where the mouse pointer's movement speed is lower than a preset threshold. When a user performs keyboard input, the system records the input characters, the start time of input, and the end time of input. The input characters include the filter conditions or search keywords entered in the filter control. When a user performs a control operation, the system records the control type, operation parameters, and changes in the interface state before and after the operation. Operation parameters include the scaling ratio change value of the scaling control and the filter condition setting value of the filter control. The recorded operations are integrated in chronological order to generate operation trajectory data that includes operation type, operation object, operation time, and operation parameters. The operation trajectory data is stored in the form of logs, and each log entry contains a unique trajectory number.

6. The data asset display and analysis method based on data visualization according to claim 3, characterized in that, The step of adjusting the multi-level display template and updating the visual interactive interface based on the operation trajectory data includes: Analyze the mouse click records in the operation trajectory data, count the click frequency of different interactive controls and data asset nodes, and identify objects with a click frequency higher than a preset click threshold as high-frequency operation objects; Adjust the display attributes of frequently accessed objects in the multi-level display template. For frequently accessed interactive controls, increase their display size on the interface and adjust their positions to easily accessible areas. For frequently accessed data asset nodes, enhance their highlighting effect in the corresponding display layer. Analyze the mouse movement records in the operation trajectory data, extract the main mouse movement path and the distribution of dwell points, and determine the user's visual attention area. The visual attention area refers to the interface area where the total dwell time exceeds a preset proportion. Adjust the layout of the multi-level display template, prioritizing the placement of core data asset nodes and important interactive controls within the visually noticeable area, while also optimizing the arrangement of nodes in the display layer; Analyze the control operation records in the operation trajectory data, extract commonly used filter conditions, zoom ratios and comparison combinations, and save them as default configurations. When the user opens the visual interactive interface again, the default configurations are automatically applied. Based on the adjusted multi-level display template, update the layout, display attributes, and default configuration of the visual interactive interface. Use smooth transition animations during the update process to complete the update of the visual interactive interface.

7. A data asset display and analysis system based on data visualization, characterized in that, The method includes a processor and a memory, the memory and the processor being connected. The memory is used to store programs, instructions or code, and the processor is used to execute the programs, instructions or code in the memory to implement the data asset display and analysis method based on data visualization as described in any one of claims 1-6.

Images