Big data-based sales data visualization method

Abstract

The invention discloses a sales data visualization method based on big data, and belongs to the technical field of data processing, and the method comprises the following steps: S1, determining the representative preference of each sales entry in a Baijiu sales report; s2, taking the position of the representative preference of each sales entry as a to-be-rendered area; s3, rendering the to-be-rendered area of each sales entry in the Baijiu sales report by using a scroll bar of the scroll container to obtain a primary visual Baijiu sales report; and S4, processing the primary visual Baijiu sales report to obtain a final visual Baijiu sales report. According to the method, the primary visual report is further updated to become the perspective report, so that the method better meets the actual requirements of the user.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and in particular relates to a sales data visualization method based on big data. Background Art

[0002] In the field of liquor sales, companies generate a large amount of sales data every day, which contains rich sales information, such as liquor sales, liquor sales channels, customer preferences and product categories, etc. Users in different places have different preferences for liquor, so it is important to effectively analyze user preferences.

[0003] In order to make more effective use of sales data and improve the scientificity and accuracy of sales decisions, sales data visualization methods based on big data have emerged. Sales report visualization processes and analyzes massive sales data and marks key information, such as the user's representative preferences recorded in the sales report. Quickly extracting the key content of the user's representative preferences can help sales staff and management quickly gain insight into sales trends, identify potential business opportunities, and optimize sales strategies.

[0004] However, most of the existing sales visualization methods have the following problems: First, the data presentation method is single and lacks interactivity; second, the data rendering effect is poor, especially for sales reports containing a large number of sales items, it is impossible to extract representative content from the user's representative preferences, and the rendering efficiency of representative content is low, affecting the user experience; third, the visualization effect is limited, and it is difficult to fully display the complexity and diversity of sales data. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a sales data visualization method based on big data.

[0006] The technical solution of the present invention is: a sales data visualization method based on big data comprises the following steps:

[0007] S1. Determine the representative preference of each sales item in the liquor sales report;

[0008] S2, taking the position of the representative preference of each sales item as the area to be rendered;

[0009] S3, using the scroll bar of the scroll container to render the to-be-rendered area of ​​each sales item in the sales report, to obtain a primary visualized liquor sales report;

[0010] S4. Process the primary visualized liquor sales report to obtain the final visualized liquor sales report.

[0011] Furthermore, S1 includes the following sub-steps:

[0012] S11, obtaining customer behavior information of each sales item in the liquor sales report;

[0013] S12, calculating the target weight of each word in the sales item according to the customer behavior information of the sales item;

[0014] S13. After sorting the target weights from large to small, the top K words are used as representative preferences for sales items, where K represents the number of words included in the preset representative preferences.

[0015] The beneficial effect of the above further scheme is: in the present invention, the target weight of each word in the customer behavior information recorded in the sales item (such as the user's purchase frequency and purchase habits and other data) is calculated, and the word with the largest weight is selected as the representative preference of the sales item, which can objectively identify the most representative information in the sales item and avoid the subjectivity and uncertainty of human selection. The introduction of the sparsity parameter in the target weight helps to deal with the vocabulary sparsity problem that may appear in the sales item, that is, some words may not be common in the sales item, but their appearance is crucial to understanding the meaning of the sales item. By adjusting the value of the sparsity parameter, the influence of common words and rare words in the weight calculation can be balanced. By using word vectors to calculate the Euclidean distance, the semantic similarity between words can be captured, and the meaning of the sales item and the customer's preference can be understood more accurately.

[0016] Furthermore, in S12, the target weight P of the i-th word in the sales item i The calculation formula is: ; where α represents the sparsification parameter, D i_NAME represents the Euclidean distance between the word vector of the i-th word in the sales item and the word vector of the product name of the sales item, D i_REPORT The Euclidean distance between the word vector of the i-th word in the sales entry and the word vector of the name of the sales report.

[0017] Furthermore, S3 includes the following sub-steps:

[0018] S31, building a scroll container;

[0019] S32, determining the monitoring duration for the scroll container to complete the virtual scrolling;

[0020] S33. During the monitoring time, the scroll bar of the scroll container is used to virtually scroll the to-be-rendered area of ​​each sales item in the sales report, and the rendering is completed to obtain a primary visualized liquor sales report.

[0021] The beneficial effect of the above further scheme is: in the present invention, virtual scrolling is a technology for optimizing the rendering performance of long lists, which reduces the number of DOM nodes and improves the scrolling response speed by rendering only the elements in the visible area. As a visual interface element, the scrolling container allows users to browse a large amount of sales data through the scroll bar, realizes the dynamic loading of data, and reduces the resources and time required for one-time loading and rendering of a large amount of data. In addition, by reasonably setting the monitoring time, it can be ensured that when the user scrolls to a new data area, there is enough time to load and render the data in the area, thereby avoiding delays or freezes in data loading. During the monitoring time, the scroll bar of the scrolling container is used to render the to-be-rendered area of ​​each sales item in the sales report.

[0022] Furthermore, in S31, the starting index K of the scroll container start The calculation formula is: ; In the formula, p y Indicates the pixel value of the vertical scroll bar of the scroll container. H represents the height of the sales item in the sales report. Represents the floor function.

[0023] Furthermore, in S32, the calculation formula of the monitoring time T of the virtual scrolling is: ; Where V max Indicates the maximum reasonable scrolling speed of the scroll container, V indicates the current scrolling speed of the scroll container, t * Indicates the basic listening interval of the scroll container.

[0024] Calculate the time from when the scroll event is triggered to when the rendering is completed as the monitoring duration. The basic monitoring interval of the scroll container usually refers to the time interval set in the scroll event monitoring.

[0025] Further, in S33, the number of buffer items N during virtual scrolling is buffer The calculation formula is: ; In the formula, min(·) represents the minimum function, L represents the width of the sales report, and M represents the length of the sales report. Represents the ceiling function.

[0026] Buffer items refer to additional rendering items outside the visible area to prevent blank spaces from appearing when scrolling quickly.

[0027] Furthermore, in S4, the Tableau tool is used to update the primary visualized liquor sales report to obtain the final visualized liquor sales report.

[0028] The beneficial effects of the present invention are:

[0029] (1) The present invention discloses a sales data visualization method based on big data, which extracts key information and trends from sales data, such as hot-selling products, sales peak periods, and customer preferences recorded in customer behavior information, and helps to quickly identify sales highlights by determining representative preferences;

[0030] (2) The present invention maps representative preferences to specific areas to be rendered, which is the key point of data visualization. The scroll bar of the scroll container is used to render the areas to be rendered of each sales item in the sales report. The dynamic rendering method can not only improve the response speed of visualization, but also accurately mark important information.

[0031] (3) The present invention further updates the primary visualization report to make it a perspective report that better meets the actual needs of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 The flowchart of the sales data visualization method based on big data. DETAILED DESCRIPTION

[0033] The embodiments of the present invention will be further described below in conjunction with the accompanying drawings.

[0034] like Figure 1 As shown, the present invention provides a sales data visualization method based on big data, comprising the following steps:

[0035] S1. Determine the representative preference of each sales item in the liquor sales report;

[0036] S2, taking the position of the representative preference of each sales item as the area to be rendered;

[0037] S3, using the scroll bar of the scroll container to render the to-be-rendered area of ​​each sales item in the liquor sales report, to obtain a primary visualized liquor sales report;

[0038] S4. Process the primary visualized liquor sales report to obtain the final visualized liquor sales report.

[0039] In this embodiment of the present invention, S1 includes the following sub-steps:

[0040] S11, obtaining customer behavior information of each sales item in the liquor sales report;

[0041] S12, calculating the target weight of each word in the sales item according to the customer behavior information of the sales item;

[0042] S13. After sorting the target weights from large to small, the top K words are used as representative preferences for sales items, where K represents the number of words included in the preset representative preferences.

[0043] In the present invention, the target weight of each word in the customer behavior information recorded in the sales item (such as the user's purchase frequency and purchase habits, etc.) is calculated, and the word with the largest weight is selected as the representative preference of the sales item, which can objectively identify the most representative information in the sales item and avoid the subjectivity and uncertainty of human selection. The introduction of the sparsity parameter in the target weight helps to deal with the vocabulary sparsity problem that may appear in the sales item, that is, some words may not be common in the sales item, but their appearance is crucial to understanding the meaning of the sales item. By adjusting the value of the sparsity parameter, the influence of common words and rare words in the weight calculation can be balanced. By using word vectors to calculate the Euclidean distance, the semantic similarity between words can be captured, and the meaning of the sales item and the customer's preference can be understood more accurately.

[0044] In the embodiment of the present invention, in S12, the target weight P of the i-th word in the sales item i The calculation formula is: ; where α represents the sparsification parameter, D i_NAME represents the Euclidean distance between the word vector of the i-th word in the sales item and the word vector of the product name of the sales item, D i_REPORT The Euclidean distance between the word vector of the i-th word in the sales entry and the word vector of the name of the sales report.

[0045] In this embodiment of the present invention, S3 includes the following sub-steps:

[0046] S31, building a scroll container;

[0047] S32, determining the monitoring duration for the scroll container to complete the virtual scrolling;

[0048] S33. During the monitoring time, the scroll bar of the scroll container is used to virtually scroll the to-be-rendered area of ​​each sales item in the sales report, and the rendering is completed to obtain a primary visualized liquor sales report.

[0049] In the present invention, virtual scrolling is a technology for optimizing the rendering performance of long lists, which reduces the number of DOM nodes and improves the scrolling response speed by rendering only the elements in the visible area. As a visual interface element, the scroll container allows users to browse a large amount of sales data through a scroll bar, realizes dynamic loading of data, and reduces the resources and time required for one-time loading and rendering of a large amount of data. In addition, by reasonably setting the monitoring time, it can be ensured that when the user scrolls to a new data area, there is enough time to load and render the data in the area, thereby avoiding delays or freezes in data loading. During the monitoring time, the scroll bar of the scroll container is used to render the area to be rendered for each sales item in the sales report.

[0050] In the embodiment of the present invention, in S31, the starting index K of the scroll container start The calculation formula is: ; In the formula, p y Indicates the pixel value of the vertical scroll bar of the scroll container. H represents the height of the sales item in the sales report. Represents the floor function.

[0051] In the embodiment of the present invention, in S32, the calculation formula of the monitoring time length T of the virtual scrolling is: ; Where V max Indicates the maximum reasonable scrolling speed of the scroll container, V indicates the current scrolling speed of the scroll container, t * Indicates the basic listening interval of the scroll container.

[0052] Calculate the time from when the scroll event is triggered to when the rendering is completed as the monitoring duration. The basic monitoring interval of the scroll container usually refers to the time interval set in the scroll event monitoring.

[0053] In the embodiment of the present invention, in S33, the number of buffer items N when performing virtual scrolling buffer The calculation formula is: ; In the formula, min(·) represents the minimum function, L represents the width of the sales report, and M represents the length of the sales report. Represents the ceiling function.

[0054] Buffer items refer to additional rendering items outside the visible area to prevent blank spaces from appearing when scrolling quickly.

[0055] In the embodiment of the present invention, in S4, the primary visualized liquor sales report is updated using the Tableau tool to obtain a final visualized liquor sales report.

[0056] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A sales data visualization method based on big data, characterized in that: The following steps are involved: S1. Determine the representative preference of each sales item in the liquor sales report; S2, taking the position of the representative preference of each sales item as the area to be rendered; S3, using the scroll bar of the scroll container to render the to-be-rendered area of ​​each sales item in the liquor sales report, to obtain a primary visualized liquor sales report; S4. Process the primary visualized liquor sales report to obtain the final visualized liquor sales report.

2. The sales data visualization method based on big data according to claim 1 is characterized in that: The S1 comprises the following sub-steps: S11, obtaining customer behavior information of each sales item in the liquor sales report; S12, calculating the target weight of each word in the sales item according to the customer behavior information of the sales item; S13. After sorting the target weights from large to small, the top K words are used as representative preferences for sales items, where K represents the number of words included in the preset representative preferences.

3. The sales data visualization method based on big data according to claim 2 is characterized in that: In S12, the target weight P of the i-th word in the sales item i The calculation formula is: ; where α represents the sparsification parameter, D i_NAME represents the Euclidean distance between the word vector of the i-th word in the sales item and the word vector of the product name of the sales item, D i_REPORT The Euclidean distance between the word vector of the i-th word in the sales entry and the word vector of the name of the sales report.

4. The sales data visualization method based on big data according to claim 1, characterized in that: The S3 comprises the following sub-steps: S31, building a scroll container; S32, determining the monitoring duration for the scroll container to complete the virtual scrolling; S33. During the monitoring time, the scroll bar of the scroll container is used to virtually scroll the to-be-rendered area of ​​each sales item in the sales report, and the rendering is completed to obtain a primary visualized liquor sales report.

5. The sales data visualization method based on big data according to claim 4 is characterized in that: In S31, the starting index K of the scroll container start The calculation formula is: ; In the formula, p y Indicates the pixel value of the vertical scroll bar of the scroll container. H represents the height of the sales item in the sales report. Represents the floor function.

6. The sales data visualization method based on big data according to claim 4 is characterized in that: In S32, the calculation formula of the monitoring time length T of the virtual scrolling is: ; Where V max Indicates the maximum reasonable scrolling speed of the scroll container, V indicates the current scrolling speed of the scroll container, t * Indicates the basic listening interval of the scroll container.

7. The sales data visualization method based on big data according to claim 4 is characterized in that: In S33, the number of buffer items N when performing virtual scrolling buffer The calculation formula is: ; In the formula, min(·) represents the minimum function, L represents the width of the sales report, and M represents the length of the sales report. Represents the ceiling function.

8. The sales data visualization method based on big data according to claim 1 is characterized in that: In S4, the primary visualized liquor sales report is updated using the Tableau tool to obtain a final visualized liquor sales report.

Images