Data visualization processing method and device, equipment and medium

Abstract

The invention relates to the technical field of artificial intelligence and the field of financial science and technology, and discloses a data visualization processing method and device, equipment and a medium, and the method comprises the steps: analyzing an obtained to-be-visualized data table, marking a dimension field and an index field corresponding to the to-be-visualized data table, disassembling the dimension field and the index field, and storing the dimension field and the index field in the to-be-visualized data table; establishing at least two corresponding combined parameter sets, generating a query data set corresponding to the combined parameter sets, and inputting the combined parameter sets into a preset behavior preference model according to a corresponding preset data feature analysis mode associated with the query data set, and generating at least two recommendation visualization charts corresponding to the preset data feature analysis mode. The technical problems that at present, data visualization tools are numerous, operation is complex, manual configuration generation is needed, time and labor are consumed, and efficiency is low are solved.

Description

Technical Field

[0001] The present invention relates to the fields of artificial intelligence technology and financial technology, and in particular to a data visualization processing method, device, equipment and medium. Background Art

[0002] With the development of science and technology, data has become an important productive force to promote business development. The process of data to business value includes data collation, processing, analysis and mining, and presentation. Data analysis is the process of using statistical and logical methods to analyze data sets in order to discover patterns, trends, and correlations to provide support for decision-making. Data analysis can be applied to various industries and scenarios, including market research, customer behavior analysis, product improvement, risk management, etc. With the explosive growth of data and the increase in the amount of information, effectively presenting data and being able to present data in the most intuitive way has become an important part of data value application.

[0003] In particular, data visualization is a technique that displays data in the form of graphics or images, which can help people understand and analyze data more intuitively. Data visualization can be applied to a variety of fields, including business intelligence, scientific research, engineering design, social sciences, etc. Data visualization is the process of displaying data through visual elements such as graphics and charts, with the aim of conveying information and trends in the data more intuitively. For complex data sets, visualization can help people quickly understand the meaning, structure, and relationship of the data. For example, a bar chart can clearly compare the size of different categories of data; a line chart can intuitively show the trend of data changes over time. Visualization can make patterns and trends hidden in the data easier to detect. For example, in a scatter plot, the distribution of data points can be observed, thereby discovering possible correlations. Presenting data in a visual form can make it easier for people from different backgrounds to understand and communicate. Whether it is a professional data analyst or a non-professional decision maker, you can get key information from a visualization chart.

[0004] At present, there are many data visualization tools, which are complex to operate and need to rely on manual configuration and generation, which is time-consuming, labor-intensive and inefficient. In addition, a large number of experienced personnel are needed to analyze business data in order to obtain useful data value. The business professionalism of personnel is required to be high, and the technical requirements for operators in the implementation are high, which is not conducive to the promotion of digitalization, thus leading to inefficient technical problems. Summary of the invention

[0005] The present invention provides a data visualization processing method, device, computer equipment and medium to solve the technical problems that current data visualization requires a large number of experienced personnel to analyze business data in order to obtain useful data value, and has high business professional and technical requirements for personnel, resulting in low efficiency.

[0006] In a first aspect, a data visualization processing method is provided, comprising:

[0007] Parsing the acquired data table to be visualized, and marking the dimension fields and indicator fields in the data table to be visualized;

[0008] Disassemble the dimension field and the indicator field to establish at least two combination parameter sets;

[0009] Generate a query data set corresponding to the combined parameter set, and associate a corresponding preset data feature analysis method with the query data set;

[0010] The combined parameter set is input into a preset behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method.

[0011] In a second aspect, a data visualization processing device is provided, comprising:

[0012] A parsing and marking module, used for parsing the acquired data table to be visualized, and marking the dimension fields and indicator fields in the data table to be visualized;

[0013] A combined parameter set module, used for disassembling the dimension field and the indicator field to establish at least two combined parameter sets;

[0014] A feature analysis module, used to generate a query data set corresponding to the combined parameter set, and associate a corresponding preset data feature analysis method with the query data set;

[0015] The visualization module is used to input the combined parameter set into the preset behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method.

[0016] In a third aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned data visualization processing method when executing the computer program.

[0017] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned data visualization processing method are implemented.

[0018] In the scheme implemented by the above-mentioned data visualization processing method, device, computer equipment and storage medium, the data table to be visualized can be obtained by parsing, and the dimension fields and indicator fields in the data table to be visualized can be marked, the dimension fields and the indicator fields can be disassembled, at least two combination parameter sets can be established, and a query data set corresponding to the combination parameter set can be generated. According to the preset data feature analysis method corresponding to the query data set, the combination parameter set is input into the preset behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method, which solves the current technical problems of numerous data visualization tools, complex operations, and the need to rely on manual configuration generation, which is time-consuming, labor-intensive, and inefficient. The dimension fields and the indicator fields are disassembled, at least two corresponding combination parameter sets are established, and visualization charts are generated according to the data feature analysis method, which does not require professional data analysts and has no professional requirements for operators, and data analysis visualization is conveniently and simply realized.

[0019] At the same time, the combined parameter set is input into the behavioral preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method, which further facilitates the operator to directly generate the required visualization chart through the visualization recommendation chart without learning professional knowledge. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0021] Figure 1 is a schematic diagram of an application environment of a data visualization processing method in one embodiment of the present invention;

[0022] Figure 2 is a flow chart of a data visualization processing method in one embodiment of the present invention;

[0023] Figure 3 is a structural schematic diagram of a data visualization processing device in one embodiment of the present invention;

[0024] Figure 4 is a schematic diagram of a structure of a computer device in one embodiment of the present invention;

[0025] Figure 5 It is another structural schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] The data visualization processing method provided by the embodiment of the present invention can be applied in Figure 1 In the application environment, the client communicates with the server through the network. The server can receive the user's data table to be visualized through the client, parse the obtained data table to be visualized, mark the dimension fields and indicator fields corresponding to the data table to be visualized, disassemble the dimension fields and the indicator fields, establish corresponding at least two combination parameter sets, generate a query data set corresponding to the combination parameter set, and input the combination parameter set into the preset behavior preference model according to the preset data feature analysis method associated with the query data set, generate at least two recommended visualization charts corresponding to the preset data feature analysis method, and return to the client. In the present invention, the data that needs to be analyzed under the business data is numerous, and the front-end business personnel, management personnel, etc., better understand the meaning of the business, and can use the data visualization processing solution to generate the required visualization results, which can improve the convenience and intelligence of data to expression, and realize automatic analysis of data, chart selection, and recommendation of visualization results. Among them, the client can be but not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices. The server can be implemented by an independent server or a server cluster composed of multiple servers. The present invention is described in detail below through specific embodiments.

[0028] See also Figure 2 As shown, Figure 2 A flow chart of a data visualization processing method provided by an embodiment of the present invention includes the following steps:

[0029] S10: parsing the acquired data table to be visualized, and marking the dimension fields and indicator fields corresponding to the data table to be visualized;

[0030] In this embodiment, the conversion of business data into graphic visualization requires parsing the acquired data table to be visualized, and marking the corresponding dimension fields and indicator fields in the data table to be visualized.

[0031] Specifically, the content of the data table to be visualized can be scanned and obtained, the header content of the data table to be visualized can be identified, the header can be further parsed, and the text field can be marked as a dimension field and the numerical field can be marked as an indicator field.

[0032] For example, a two-dimensional data table first implements automatic marking and differentiation of dimension and indicator columns through content scanning. This embodiment may adopt a method of data table format scanning to mark columns with text format as dimensions and columns with numerical content as indicators.

[0033] The aforementioned data table format scanning may be data table text and numerical value recognition and classification, which generally refers to the process of using OCR technology to recognize text and numerical values ​​in a data table and classify them according to their characteristics.

[0034] In the data table text recognition classification, OCR technology can recognize and extract text information in the data table. For example, OCR technology can be used to identify the text area in the data table and convert the detected text area into a corresponding character sequence. In addition, deep learning technologies such as convolutional neural networks (CNN) and recurrent neural networks (RNN) can also be used to improve the recognition accuracy. Numerical recognition classification focuses on identifying numerical data in the data table and classifying them according to their numerical size or other features. When processing two-dimensional data tables, OCR technology can effectively assist in the automatic marking and differentiation of dimensions and indicator columns. For example, image preprocessing: First, the scanned data table image needs to be preprocessed to improve the recognition accuracy. This usually includes operations such as denoising, binarization, and rotation correction. Text detection: The text area in the image is identified through a text detection algorithm, such as a CRNN network based on deep learning. Text recognition: Using an OCR engine, such as a recognition model based on deep learning, the detected text area is converted into machine-readable text. Feature extraction: Feature extraction of the recognized text, such as font size, shape, position, etc., as well as semantic analysis of the text content. Classification: Classify the text into dimensions or indicators based on the extracted features and content. In this way, columns with text format are marked as dimensions, and columns with numerical content are marked as indicators. Post-processing: Correct the recognition results, such as correcting recognition errors, formatting output, etc.

[0035] It is understandable that parsing the obtained data table to be visualized and marking the dimension fields and indicator fields corresponding to the data table to be visualized also includes: obtaining the data table to be visualized and standardizing the data format and table format of the data table to be visualized.

[0036] Specifically, standardizing the data format and table format of the data table to be visualized, such as standardizing the data format and table format of a two-dimensional data table, is an important part of data preprocessing, aiming to ensure the consistency and comparability of data. The following are some commonly used data standardization methods and steps: Data type normalization: Ensure that all data in the data table are in a unified format, such as converting all numerical data into floating point format. Decimal place normalization: Determine the number of digits after the decimal point in the data table to maintain data consistency. Format normalization: Including date format, time format, currency symbol, etc., to ensure that the date, time and currency values ​​in the data table have a unified format. Encoding conversion: For example, converting "male / female" in text format to "0 / 1" in numerical format. Missing value processing: Determine how to handle missing values ​​in the data table, such as filling the mean, median or using a prediction model to estimate. Outlier processing: Identify and handle outliers, such as determining and correcting them through the standard deviation method or box plot. Unit unification: Ensure that all data use the same measurement unit. The standardized data table should follow certain design specifications, such as the paradigm theory in relational database design, to reduce data redundancy and improve data integrity. For example, the first normal form (1NF) requires atomicity of fields, the second normal form (2NF) requires no partial dependency, and the third normal form (3NF) requires no dependency transitivity.

[0037] Among them, data type normalization is a key step to ensure data consistency and comparability so that data can be correctly displayed and analyzed during visualization. For example, unify data types: make sure all numerical columns are of the same data type (for example, floating point numbers). Text data should be encoded consistently (such as UTF-8). Normalize text data: make sure there are no spelling errors or extra spaces in text fields. For categorical data, make sure that the name of each category is consistent (for example, do not mix "Male" and "male", do not mix the full name or abbreviation of the city, etc., which are not specifically limited here). Handle missing values: determine whether to delete rows or columns with missing values ​​or fill missing values ​​(you can use the mean, median, mode, or other relevant values). Convert dates and times: make sure that all date and time data follow the same format, such as the ISO8601 standard (YYYY-MM-DD). Unify currencies and units of measurement: if the data table contains currency, make sure all amounts are converted to a unified currency unit (such as US dollars). Make sure all units of measurement are consistent (such as using meters instead of centimeters). Handle outliers: identify and decide how to deal with outliers, which may include deleting, replacing, or retaining them. Create consistent categories: For ordered categories (such as ratings), ensure that each category has a consistent representation. Convert ratio data: For ratio or percentage data, ensure that they are all expressed with the same cardinality (such as 100%). Encode categorical variables: Encode categorical variables using an appropriate method (such as one-hot encoding or label encoding). Data table structure: Ensure that the data table is two-dimensional, that is, each cell contains only one value, and each value is an instance of the same variable. Data cleaning: Remove unnecessary rows, columns, or duplicate records. Data sorting: Sort the data as needed to show a specific order or pattern in the visualization. Data stratification: If the visualization requires it, the data can be divided into multiple subsets or levels.

[0038] S11: disassemble the dimension field and the indicator field, and establish at least two corresponding combination parameter sets;

[0039] Therefore, it can be understood that when accessing a data table, the field type in the data is read first, and the text type field is used as the latitude field and the numeric type field is used as the indicator field by default.

[0040] It should be understood that in this embodiment, the dimension fields and the indicator fields are disassembled, and associated expression groups are established. Each expression group includes combining at most two dimension fields with at most two indicator fields, and at least two expression groups are combined as the combination parameter set. When there are multiple dimension columns, the expression group will be generated by dimension combination + indicator. The combinations that can be generated are, for example, "province + business volume", "province + growth rate", or the expression combinations that can be generated are "province + business volume", "city + business volume", "province + city + business volume"... When there are multiple dimension columns, it is mainly through exhaustive combination to list as many combinations as possible. It should be noted that each latitude field and indicator field are exhaustively combined to generate a corresponding calculation query parameter set. The number of latitude combinations is controlled here, and the combination does not exceed 2 latitudes and 2 indicators. Otherwise, data diffusion will occur due to too many combinations, resulting in waste of resources.

[0041] S12: Generate a query data set corresponding to the combined parameter set, and associate a corresponding preset data feature analysis method with the query data set;

[0042] After combining at least two of the expression groups into the combined parameter set, the combined parameter set is used to generate the corresponding query data set through a grouping aggregation function, and the query data set includes indicator data features corresponding to the dimension field and the indicator field; an analysis method keyword corresponding to the indicator data feature is generated, and the corresponding preset data feature analysis method is associated.

[0043] For example, the latitude and index parameters are used as parameter groups to generate corresponding data query statements, mainly performing aggregation function processing on the latitude parameters and indicators. Finally, the data set to be presented in the chart is queried from the original data table.

[0044] The aforementioned grouping and aggregation functions refer to functions used to group and aggregate data in database queries or data processing. The following functions can calculate the grouped data set and return a summary value: SUM: calculates the sum of a group of values. For example: SELECT group_column, SUM(numeric_column) FROM table GROUPBY group_column;

[0045] AVG: Calculates the average value of a group of values. For example: SELECT group_column,AVG(numeric_column) FROM table GROUP BY group_column;

[0046] COUNT: Counts the number of items in a group. For example: SELECT group_column,COUNT(*)FROMtable GROUP BY group_column;

[0047] MIN: Find the minimum value in a group of numeric values. For example: SELECT group_column,MIN(numeric_column) FROM table GROUP BY group_column;

[0048] MAX: Find the maximum value in a group of values. For example: SELECT group_column,MAX(numeric_column) FROM table GROUP BY group_column;

[0049] STDDEV or STDDEV_POP: Calculates the overall standard deviation of a group of values. For example: SELECT group_column,STDDEV(numeric_column) FROM table GROUP BY group_column;

[0050] VAR or VAR_POP: Calculates the overall variance of a set of numeric values. For example: SELECT group_column, VAR (numeric_column) FROM table GROUP BY group_column;

[0051] COUNTDistinct: Counts the number of distinct values ​​in a group. For example: SELECT group_column, COUNT(DISTINCT numeric_column) FROM table GROUP BY group_column;

[0052] GROUP_CONCAT: Concatenates a group of text values ​​into a string. For example: SELECT group_column,GROUP_CONCAT(distinct text_column ORDER BY text_column SEPARATOR',') FROM table GROUP BY group_column;

[0053] STRING_AGG: concatenates a set of text values ​​into a string. For example: SELECT group_column,STRING_AGG(text_column,',')FROM table GROUP BY group_column;

[0054] LISTAGG: Concatenates a set of text values ​​into a string. For example: SELECT group_column,LISTAGG(text_column,',')WITHIN GROUP(ORDER BY text_column)FROM table GROUP BY group_column;

[0055] It should be further explained that, according to the characteristics of the indicator data, keywords of the analysis method are generated, such as distribution, sum, comparison, average, etc. At this time, it is necessary to analyze the characteristics of the data. For example, if it involves integers or decimals of quantities, distribution, sum, comparison, maximum, minimum, etc. can be generated, and if it involves rates, trend, comparison, distribution, maximum, minimum, etc. are generated. Conventional analysis methods are sorted out, associated with expression groups, and used as input for graphic selection queries.

[0056] The step of generating an analysis method keyword corresponding to the indicator data feature and associating the corresponding preset data feature analysis method comprises:

[0057] Constructing a chart expression tag corresponding to the combined parameter set and associating it to a preset corresponding chart;

[0058] The chart expression tag and the preset corresponding chart are determined according to the analysis method keyword, and are associated with the corresponding preset data feature analysis method.

[0059] Construct common chart parameter number set expression labels and associate them with corresponding charts, such as latitude + index, which can be expressed by pie chart, bar chart, 2 latitude + index, bar chart, latitude plus 2 index, combination chart, etc.

[0060] The expression labels and graph association methods include but are not limited to: Bar Chart Parameters: Dataset, Category Label (X-axis), Numeric Label (Y-axis) Expression Label: Bar charts are usually used to compare the number of different categories. Association: Category labels correspond to the X-axis, and numerical data correspond to the Y-axis. Line Chart Parameters: Dataset, Time Series or Ordered Categories (X-axis), Numeric Label (Y-axis) Expression Label: Line charts are suitable for showing the trend of data changes over time. Association: Time or category corresponds to the X-axis, and numerical data corresponds to the Y-axis. Pie Chart Parameters: Dataset, Category Label, Numeric Label; Expression Label: Pie charts are used to show the proportion of each part to the whole. Association: Category labels correspond to each part of the pie chart, and numerical labels determine the size of each part. Scatter Plot Parameters: Dataset, Numeric Label (X-axis and Y-axis) Expression Label: Scatter charts are used to show the relationship between two numerical variables. Association: Numeric data corresponds to the X-axis and Y-axis. Histogram parameters: Dataset, Numerical Label (X-axis), Bins Expression Label: Histogram is used to show the distribution of numerical data. Association: Numerical data corresponds to the X-axis, and the bins determine the width of the histogram. Box Plot parameters: Dataset, Classification Label (X-axis), Numerical Label (Y-axis); Expression Label: Box Plot is used to show the distribution of data, including median, quartiles, and outliers. Association: Classification label corresponds to the X-axis, and numerical data corresponds to the Y-axis. Heatmap parameters: Dataset, Classification Label (X-axis and Y-axis), Color Intensity (indicates the size of the value) Expression Label: Heatmap is used to show the strength of the relationship between two categorical variables and a numerical variable. Association: Classification label corresponds to the X-axis and Y-axis, and color intensity indicates the size of the value. Area Chart parameters: Dataset, Time Series or Ordered Categories (X-axis), Numerical Label (Y-axis) Expression Label: Area chart is similar to line chart, but the area is filled with color or texture to show the cumulative effect of quantity over time. Association: Time or category corresponds to the X-axis, and numerical data corresponds to the Y-axis. Treemap parameters: data set, classification label, numerical label expression label: Treemap is used to display the size of each part in the hierarchy. Association: Classification labels correspond to each branch of the treemap, and numerical labels determine the area size of each branch. Bubble Chart parameters: data set, numerical labels (X-axis and Y-axis), bubble size (representing the third numerical variable) expression label: Bubble chart is a variation of scatter chart, adding a third numerical variable to indicate the size of the bubble. Association: Numerical data corresponds to the X-axis and Y-axis, and the third numerical variable determines the size of the bubble. Radar Chart parameters: data set, multiple classification labels (representing different variables), numerical labels. Expression label: Radar chart is used to display the numerical size of multiple variables.Association: The category label corresponds to the axis of the radar chart, and the numerical data determines the length of the radiation from the center point. Stacked Bar / AreaChart parameters: Dataset, Category Label (X-axis), Numeric Label (Y-axis), Grouping Variable (for stacking). Expression Label: Stacked charts are used to show the contribution of each part to the total. Association: The category label corresponds to the X-axis, and the numerical data of different groups are stacked on the Y-axis.

[0061] S13: Inputting the combined parameter set into a preset behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method.

[0062] After determining the chart expression label and the preset corresponding chart according to the analysis method keyword and associating them to the corresponding preset data feature analysis method, it is necessary to input the combined parameter set into the preset behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method.

[0063] Due to the combination of dimensions and indicators, there are many combinations of data sets and corresponding charts. When there are too many expression combinations, further learning and optimization of sorting and recommendation are required. That is, after generating a list of recommended charts, for example, a default list of 6 display charts is given for the first time according to the default dimension order. After that, the corresponding recommendation algorithm needs to be learned based on the recorded user behavior preferences.

[0064] The aforementioned behavior preference model, for example, when a user clicks on a chart, the number of clicks is recorded, and then the ranking is optimized based on the number of times and duration of clicks on the chart by the user, the preference coefficient value between the chart and the user is improved, and the sorting list is regenerated based on the size of the preference coefficient value. The calculation method of the preference coefficient is:

[0065] Preference coefficient = (number of clicks / maximum number of clicks*0.7) + (browsing time / maximum browsing time)*0.3, where the weight 0.7 or 0.3 can be optimized and adjusted accordingly based on different users using machine learning methods.

[0066] Specifically, it can be understood that behavioral preference models are usually used to describe and predict the behavioral choices of individuals in specific situations. When constructing a behavioral preference model, many factors are considered, including individual preferences, the costs and benefits of choices, and the influence of environmental and social factors. The following are some parameters and expression labels that may be used when constructing a behavioral preference model: Preferences: Describe the degree of preference of an individual for different choices. Expression label: Preferences can be obtained directly through surveys or observations, or they can be set through model assumptions. Utility: Quantify the degree of satisfaction that an individual obtains from different choices. Expression label: Utility values ​​are usually calculated through utility functions, which can be linear or nonlinear. Costs: Includes monetary costs, time costs, effort costs, etc. Expression label: Costs usually appear in the form of negative utility in the model. Benefits: Includes direct benefits and indirect benefits, such as money, health, happiness, etc. Expression label: Benefits are reflected in the form of positive utility in the model. Probabilities: Describe the probability of different events. Expression label: Probability is usually related to uncertainty and risk attitude. Constraints: Describes the constraints that an individual faces when making decisions. Expression Label: Constraints can be budget constraints, time constraints, or other resource constraints. Social Influence: Describes the impact of the social environment on individual behavior. Expression Label: Social influence can be reflected through social networks, group norms, or social identity. Psychological Factors: Describes the impact of an individual's psychological state on behavior. Expression Label: Psychological factors may include emotions, attitudes, personality traits, etc. Habits: Describes an individual's tendency to certain behavioral patterns due to habit. Expression Label: Habits can be measured by frequency, persistence, or degree of automaticity. Contextual Factors: Describes the factors that influence behavior in a specific situation. Expression Label: Contextual factors may include time, place, environmental atmosphere, etc.

[0067] In practical applications, behavioral preference models can be used in many fields such as market research, user profile construction, recommendation systems, health behavior change, policy making, etc. For example, in user profile construction, the user's behavior data can be used to calculate their preferences, thereby providing personalized service or product recommendations.

[0068] It should be noted that the aforementioned recommendation algorithms can filter out the most valuable information from a large amount of data in data analysis. The following are some commonly used recommendation algorithms and their application scenarios:

[0069] Collaborative filtering recommendation algorithm: User-based collaborative filtering: recommends items by analyzing the similarities between users, which is suitable for scenarios with a large user group. Item-based collaborative filtering: recommends items by analyzing the similarities between items, which is suitable for scenarios with a rich variety of items. Content-based recommendation algorithm: recommends similar items based on the characteristics of items and the historical behavior of users, which is suitable for content recommendations such as news, articles, and videos. Matrix decomposition algorithm: predicts the user's rating of unrated items by decomposing the user-item rating matrix, which is suitable for situations where rating data is sparse. Deep learning recommendation model: uses deep neural networks to learn the implicit feature representation of users and items to improve the accuracy of recommendations, which is suitable for large-scale data sets. Hybrid recommendation algorithm: combines multiple recommendation algorithms to improve the accuracy and coverage of recommendations, which is suitable for diverse user needs. Group recommendation model: considers the preferences of group members and the overall preferences of the group, which is suitable for group decision-making scenarios such as tourism and catering. Attention-based group representation learning: uses the attention mechanism to learn the weights of group members, which is suitable for scenarios where the influence of members within the group needs to be considered. Neural collaborative filtering (NCF): embeds users and items into neural networks to learn interaction functions, which is suitable for scenarios that need to capture complex user-item interactions.

[0070] In practical applications, the selection and design of recommendation algorithms should be determined according to specific business needs and data characteristics. For example, when building a movie data analysis recommendation system, a content-based recommendation algorithm may be used to recommend new movies similar to the movies that users have previously rated highly; while in an e-commerce platform, a collaborative filtering algorithm may be used to recommend products purchased by other users who have purchased similar products, and finally visualize all data sets.

[0071] The aforementioned re-generated sorted list is generated according to the size of the preference coefficient value, so as to present the most suitable chart at the front as much as possible, realize the recommendation model based on usage habits, and improve the user experience.

[0072] The data visualization processing method provided in this embodiment can be obtained by parsing the data table to be visualized, and marking the dimension fields and indicator fields corresponding to the data table to be visualized, disassembling the dimension fields and the indicator fields, establishing at least two corresponding combination parameter sets, generating a query data set corresponding to the combination parameter set, and inputting the combination parameter set into the preset behavior preference model according to the preset data feature analysis method associated with the query data set, generating at least two recommended visualization charts corresponding to the preset data feature analysis method, solving the technical problems that there are many data visualization tools, the operation is complicated, and the generation depends on manual configuration, which is time-consuming and labor-intensive, and the efficiency is low. The dimension fields and the indicator fields are disassembled, at least two corresponding combination parameter sets are established, and visualization charts are generated according to the data feature analysis method, realizing that professional data analysts are not required, and there are no professional requirements for operators, and data analysis visualization is realized conveniently and simply. At the same time, the combination parameter set is input into the behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method, which further facilitates the operator to directly generate the required visualization chart through the visualization recommendation chart without learning professional knowledge.

[0073] It can also realize the transformation from data tables to business relationship expression combinations. It provides a foundation for data intelligence and also provides the underlying foundation for subsequent text questions to data presentation. It can more effectively transform data into charts and improve the business's understanding of data. It can improve the efficiency of data analysis and give full play to the value of data. It can reduce the workload of manually making tables and using BI to analyze data in the past, reduce costs, improve efficiency, and have lower requirements for professional capabilities. It will also provide a foundation for subsequent intelligent autonomous analysis and the generation of analytical texts. It can reduce the workload of daily writing and reporting.

[0074] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0075] In one embodiment, a data visualization processing device is provided, and the data visualization processing device corresponds one-to-one to the data visualization processing method in the above embodiment. Figure 3 As shown, the data visualization processing device includes a parsing and marking module 101, a combined parameter set module 102, a feature analysis module 103, and a visualization module 104. The functional modules are described in detail as follows:

[0076] A parsing and marking module 101 is used to parse the acquired data table to be visualized, and mark the dimension fields and indicator fields corresponding to the data table to be visualized;

[0077] A combined parameter set module 102, used to decompose the dimension field and the indicator field to generate at least two corresponding combined parameter sets;

[0078] The feature analysis module 103 is used to generate a query data set corresponding to the combined parameter set, and associate a corresponding preset data feature analysis method with the query data set;

[0079] The visualization module 104 is used to input the combined parameter set into the preset behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method.

[0080] In one embodiment, the data visualization processing device further includes:

[0081] A calculation module, used to calculate the preference coefficient corresponding to the pre-recorded user behavior preference using a preset recommendation algorithm;

[0082] A preference model module, used for establishing the preset behavior preference model according to the preference coefficient;

[0083] An acquisition module, used for acquiring the data table to be visualized;

[0084] The standardization module is used to standardize the data format and table format of the data table to be visualized.

[0085] In one embodiment, the parsing tag module 101 is specifically used to:

[0086] Scan the obtained content of the data table to be visualized, and identify the header content of the data table to be visualized;

[0087] The table header is parsed, marking text fields as dimension fields and numeric fields as indicator fields.

[0088] In one embodiment, the combined parameter set module 102 is specifically configured to:

[0089] Decomposing the dimension fields and the indicator fields, and establishing associated expression groups, each of the expression groups including at most two dimension fields and at most two indicator fields;

[0090] At least two of the expression groups are combined into the combined parameter set.

[0091] In one embodiment, the feature analysis module 102 is specifically configured to:

[0092] The combined parameter set is used to generate the corresponding query data set through a grouping aggregation function, where the query data set includes indicator data features corresponding to the dimension field and the indicator field;

[0093] Generate an analysis method keyword corresponding to the indicator data feature, and associate the corresponding preset data feature analysis method.

[0094] The feature analysis module 102 specifically generates analysis method keywords corresponding to the indicator data features and associates the corresponding preset data feature analysis methods for:

[0095] Constructing a chart expression tag corresponding to the combined parameter set and associating it to a preset corresponding chart;

[0096] The chart expression tag and the preset corresponding chart are determined according to the analysis method keyword, and are associated with the corresponding preset data feature analysis method.

[0097] The data visualization processing device provided in this embodiment can analyze the obtained data table to be visualized, mark the dimension fields and indicator fields corresponding to the data table to be visualized, disassemble the dimension fields and the indicator fields, establish at least two corresponding combination parameter sets, generate a query data set corresponding to the combination parameter set, and input the combination parameter set into the preset behavior preference model according to the preset data feature analysis method associated with the query data set, and generate at least two recommended visualization charts corresponding to the preset data feature analysis method, which solves the technical problems that there are many data visualization tools, the operation is complicated, and the generation depends on manual configuration, which is time-consuming and labor-intensive, and the efficiency is low. The dimension fields and the indicator fields are disassembled, at least two corresponding combination parameter sets are established, and visualization charts are generated according to the data feature analysis method, which realizes that professional data analysts are not required, and there are no professional requirements for operators, and data analysis visualization is realized conveniently and simply. At the same time, the combination parameter set is input into the behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method, which further facilitates the operator to directly generate the required visualization chart through the visualization recommendation chart without learning professional knowledge.

[0098] It can also realize the transformation from data tables to business relationship expression combinations. It provides a foundation for data intelligence and also provides the underlying foundation for subsequent text questions to data presentation. It can more effectively transform data into charts and improve the business's understanding of data. It improves the efficiency of data analysis and maximizes the value of data. It reduces the workload of manually making tables and using BI to analyze data in the past, reduces costs, improves efficiency, and requires lower professional capabilities. It will also provide a foundation for subsequent intelligent autonomous analysis and the generation of analytical texts. It reduces the workload of daily writing and reporting.

[0099] For the specific definition of the data visualization processing device, please refer to the definition of the data visualization processing method above, which will not be repeated here. Each module in the above-mentioned data visualization processing device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0100] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client via a network connection. When the computer program is executed by the processor, it implements the functions or steps of a data visualization processing method server side.

[0101] In one embodiment, a computer device is provided. The computer device may be a client, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server through a network connection. When the computer program is executed by the processor, it realizes the functions or steps of the client side of a data visualization processing method.

[0102] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0103] Parse the acquired data table to be visualized, and mark the dimension fields and indicator fields corresponding to the data table to be visualized;

[0104] Disassemble the dimension field and the indicator field, and establish at least two corresponding combination parameter sets;

[0105] Generate a query data set corresponding to the combined parameter set, and associate a corresponding preset data feature analysis method with the query data set;

[0106] The combined parameter set is input into a preset behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method.

[0107] The computer device provided in this embodiment can parse the obtained data table to be visualized, mark the dimension fields and indicator fields corresponding to the data table to be visualized, disassemble the dimension fields and the indicator fields, establish at least two corresponding combination parameter sets, generate a query data set corresponding to the combination parameter set, and input the combination parameter set into the preset behavior preference model according to the preset data feature analysis method associated with the query data set, and generate at least two recommended visualization charts corresponding to the preset data feature analysis method, which solves the technical problems that there are many data visualization tools, the operation is complicated, and the generation depends on manual configuration, which is time-consuming and labor-intensive, and the efficiency is low. The dimension fields and the indicator fields are disassembled, at least two corresponding combination parameter sets are established, and visualization charts are generated according to the data feature analysis method, which realizes that professional data analysts are not required, and there are no professional requirements for operators, and data analysis visualization is realized conveniently and simply. At the same time, the combination parameter set is input into the behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method, which further facilitates the operator to directly generate the required visualization chart through the visualization recommendation chart without learning professional knowledge.

[0108] It can also realize the transformation from data tables to business relationship expression combinations. It provides a foundation for data intelligence and also provides the underlying foundation for subsequent text questions to data presentation. It can more effectively transform data into charts and improve the business's understanding of data. It improves the efficiency of data analysis and maximizes the value of data. It reduces the workload of manually making tables and using BI to analyze data in the past, reduces costs, improves efficiency, and requires lower professional capabilities. It will also provide a foundation for subsequent intelligent autonomous analysis and the generation of analytical texts. It reduces the workload of daily writing and reporting.

[0109] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0110] Parse the acquired data table to be visualized, and mark the dimension fields and indicator fields corresponding to the data table to be visualized;

[0111] Disassemble the dimension field and the indicator field, and establish at least two corresponding combination parameter sets;

[0112] Generate a query data set corresponding to the combined parameter set, and associate a corresponding preset data feature analysis method with the query data set;

[0113] The combined parameter set is input into a preset behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method.

[0114] The computer-readable storage medium provided in this embodiment can be obtained by parsing the data table to be visualized, and marking the dimension fields and indicator fields corresponding to the data table to be visualized, disassembling the dimension fields and the indicator fields, establishing at least two corresponding combination parameter sets, generating a query data set corresponding to the combination parameter set, and inputting the combination parameter set into the preset behavior preference model according to the preset data feature analysis method associated with the query data set, generating at least two recommended visualization charts corresponding to the preset data feature analysis method, solving the technical problems that there are many data visualization tools, complex operations, and need to rely on manual configuration generation, which is time-consuming and labor-intensive, and inefficient. The dimension fields and the indicator fields are disassembled, at least two corresponding combination parameter sets are established, and visualization charts are generated according to the data feature analysis method, realizing that professional data analysts are not required, and there are no professional requirements for operators, and data analysis visualization is realized conveniently and simply. At the same time, the combination parameter set is input into the behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method, which further facilitates the operator to directly generate the required visualization chart through the visualization recommendation chart without learning professional knowledge.

[0115] It can also realize the transformation from data tables to business relationship expression combinations. It provides a foundation for data intelligence and also provides the underlying foundation for subsequent text questions to data presentation. It can more effectively transform data into charts and improve the business's understanding of data. It improves the efficiency of data analysis and maximizes the value of data. It reduces the workload of manually making tables and using BI to analyze data in the past, reduces costs, improves efficiency, and requires lower professional capabilities. It will also provide a foundation for subsequent intelligent autonomous analysis and the generation of analytical texts. It reduces the workload of daily writing and reporting.

[0116] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can refer to the relevant descriptions on the server side and the client side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0117] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0118] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0119] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

[0120] It should be noted that if software tools or components other than those of our company appear in the embodiments of the present application, they are only used for illustrative purposes and do not represent actual use.

Claims

1. A data visualization processing method, characterized in that: include: Parsing the acquired data table to be visualized, and marking the dimension fields and indicator fields in the data table to be visualized; Disassemble the dimension field and the indicator field to establish at least two combination parameter sets; Generate a query data set corresponding to the combined parameter set, and associate a corresponding preset data feature analysis method with the query data set; The combined parameter set is input into a preset behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method.

2. The data visualization processing method according to claim 1, characterized in that: The method of parsing the obtained data table to be visualized and marking the dimension fields and indicator fields in the data table to be visualized includes: Scan the obtained content of the data table to be visualized, and identify the header content of the data table to be visualized; The table header is parsed, text fields are marked as dimension fields, and numeric fields are marked as indicator fields.

3. The data visualization processing method according to claim 1, characterized in that: The disassembling of the dimension field and the indicator field to establish at least two combination parameter sets includes: Decomposing the dimension fields and the indicator fields, and establishing associated expression groups, each of the expression groups including at most two dimension fields and at most two indicator fields; At least two of the expression groups are combined into the combined parameter set.

4. The data visualization processing method according to claim 1, characterized in that: The generating of a query data set corresponding to the combined parameter set and associating a corresponding preset data feature analysis method with the query data set includes: Generate a query data set by using a grouping aggregation function on the combined parameter set, wherein the query data set includes indicator data features corresponding to the dimension field and the indicator field; Generate analysis method keywords for the indicator data features and associate them with corresponding preset data feature analysis methods.

5. The data visualization processing method according to claim 4, characterized in that: The generating of the analysis method keywords corresponding to the indicator data features and associating the corresponding preset data feature analysis methods includes: Constructing a chart expression label corresponding to the combined parameter set and associating it to a preset corresponding chart; The chart expression tag and the preset corresponding chart are determined according to the analysis method keyword, and are associated with the corresponding preset data feature analysis method.

6. The data visualization processing method according to claim 1, characterized in that: The step of inputting the combined parameter set into a preset behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method may also include: A preset recommendation algorithm is used to calculate the preference coefficient corresponding to the pre-recorded user behavior preference; The preset behavior preference model is established according to the preference coefficient.

7. The data visualization processing method according to claim 1, characterized in that: The step of parsing the obtained data table to be visualized and marking the dimension fields and indicator fields corresponding to the data table to be visualized also includes: Obtain the data table to be visualized; The data format and table format of the data table to be visualized are standardized.

8. A data visualization processing device, characterized in that: include: A parsing and marking module, used for parsing the acquired data table to be visualized, and marking the dimension fields and indicator fields in the data table to be visualized; A combined parameter set module, used for disassembling the dimension field and the indicator field to establish at least two combined parameter sets; A feature analysis module, used to generate a query data set corresponding to the combined parameter set, and associate a corresponding preset data feature analysis method with the query data set; The visualization module is used to input the combined parameter set into the preset behavior preference model to generate at least two recommended visualization charts corresponding to the preset data feature analysis method.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the data visualization processing method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the data visualization processing method according to any one of claims 1 to 7 are implemented.

Images