Shulou(Shulou.com)06/01 Report--

This article introduces you how to use QuickBI to build cool visual analysis, the content is very detailed, interested friends can refer to, hope to be helpful to you.

With the penetration of big data in various industries, the demand for BI data analysis is increasing day by day. How to make visualization better show the value of data is the direction of BI products. Domestic and foreign BI products have their own methods, such as foreign brands PowerBI, Tableau, as well as domestic FineBI, BDP, Quick BI they all provide richness visualization capabilities, but for beginners, how to choose the appropriate chart after having the data? How to configure cool and dazzling charts? How do multiple charts organize reports with stories?

Chart development

To use Quick BI for visual analysis, you only need three simple steps (after you open the trial, you can go to the link below, click here to start the trial)

1. Connect to the data source and open the http://bi.aliyun.com/workspace/datasource easy configuration. At present, Quick BI already supports most data source types, including Mysql/PostgreSQL/SQL Server/Hive

two。 Create a new dataset and open http://bi.aliyun.com/workspace/schema to select a table in the database to create a dataset. The dataset also supports snowflake model and star model association between multiple tables.

3. Create the dashboard, and then you can start selecting the chart you want.

First, take a look at the types of Quick BI charts:

As shown above, Quick BI currently provides 13 categories and more than 30 kinds of charts. For example, bar charts include ordinary bar charts, stacked bar charts, percentage stacked bar charts, and percentage stacked bar charts, which can cover the vast majority of scenarios in BI analysis.

Quick BI can also easily build line-column combination diagrams. As shown in the following figure, in order to analyze the relationship between clothing category profit and the number of visitors and sales, the use of profit column display is helpful to highlight the primary and secondary relationship.

Mouse hover to the legend can highlight the specified data, suitable for a single dimension analysis in the case of too many dimensions.

In addition, compared with other charts, Quick BI adapts to scenarios with large amounts of data to avoid copy coverage or incomplete presentation. As shown in the following figure, when there are too many use cases of the pie chart, the scroll bar display will be automatically enabled for the legend, and the Tooltip display position will be optimized:

In addition, there is a source direction graph, which is suitable for analyzing page-to-page adjustments and user behavior paths:

QuickBI provides so many ways to display charts, but how to choose the right chart type? This needs to start according to the characteristics of the data, to master the 1st Skill needs practice and experience, for beginners, you can refer to the following chart to select the flow chart.

Refer to Dr. Andrew Abela's chart suggestion (http://extremepresentation.typepad.com/blog/2006/09/choosing_a_good.html)

Principle of chart realization

After introducing the type of chart and how to use it, if you are interested in the principle of chart technology, you can see the following chart implementation principle analysis.

The chart framework is divided into four layers from top to bottom:

1. Component layer

Based on the different ways of displaying the chart, the interaction is very different, so the line graph, column chart, area chart, bar chart, pie chart and so on are returned to the basic chart, which is constructed based on the conventional Cartesian coordinate system or polar coordinate system. Cross table, matrix tree chart, funnel chart, source direction graph and other interactions are different, which need to be optimized and classified as rich interactive charts. The map needs to be based on LBS geographic location data. At present, there are built-in maps of Chinese provinces, cities and counties, which can be drilled up and down at different levels. The last category is the 3D chart of the future plan to launch. Combined with the current rise of VR/AR devices, 3D charts have a better visual experience, and plans to make efforts in the future.

two。 Chart interaction layer

Static charts are rigid, and flexible interactions improve the efficiency and experience of data analysis just like making charts speak. Drilling linkage jump is a basic interaction for OLAP data, and all these supporting classes of charts can be implemented uniformly. Axis Axis, legend, prompt box Tooltip are supported in multiple charts, packaged as a general module. For a large amount of data, these three general modules need to solve the problem of how to display a large number of copywriting in a small space. Quick BI internal display optimization based on many algorithms, such as when the number of axes is too large, by comparing the axis width and the copy width, automatically calculate the axis copy tilt angle to avoid copy overlap. For the case of excessive data, automatic sampling display is realized.

3. Data configuration layer

Data is mainly divided into two categories: "presentation data" and "chart configuration data".

The presentation data is the original data in the business and will not change according to the type of chart.

Chart configuration data is divided into "user-generated chart configuration" and "chart default configuration". The final chart display is based on the combination of the two.

Because the upper chart interaction is only responsible for presentation, the data structure needed is also for efficient display, and the data provided by the interface is for convenient storage, so the two are inconsistent in many cases, so it is necessary for the data conversion layer to adapt the data format.

This layer is the core of the whole data processing, not only the conversion method, but also the design of the data structure, which needs to take into account the horizontal and vertical considerations of different chart types.

4. Bottom dependence

Due to the diversity of visual charts, a set of framework or drawing method is difficult to meet business needs efficiently, so the bottom layer is based on G2, Three.js, Leaflet three basic libraries, these three are not ready-made chart libraries, all need to implement charts. Similar to the relationship between flour and noodles, these three are equivalent to different tastes of flour, and the final chart is like noodles.

G2 is a set of graphic syntax based on visual coding, data-driven and suitable for basic charts; Three.js is the encapsulation of WebGL and provides efficient API for developing 3D charts; Leaflet is suitable for the development of interactive map charts

On how to use QuickBI to build cool visual analysis to share here, I hope the above content can be of some help to you, can learn more knowledge. If you think the article is good, you can share it for more people to see.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.