Data Matching and Searching

1. Overview

1.1 Application scenario

During data processing, the data of multiple data tables are often scattered. Sometimes, users want to realize data matching and search between data tables through a field (primary key) or multiple fields (joint primary key) between data tables.

For example, there are two data tables: "Internet_User information table" and "Internet_ Access statistics table"

"Internet_User information table" only stores the user ID and does not know the access time, number of access, residence time and other information of the corresponding user. You can use "left and right merge" to correlate and find data, and merge the data into one data table to facilitate subsequent analysis of user access data.

Note: the user ID is not repeated.

1.2 function introduction

In BI, self-service data sets can use "left and right merge", similar to the VLOOKUP function in Excel.

2. Operation steps

Sample data:

408.Sample data.rar

2.1 Create self-service dataset

Upload the dataset.

Create a self-service dataset, as shown in the following figure:

2.2 Match the field

Select "Internet_User information table", as shown in the following figure:

Create a "left and right merge" step, select"Internet_User information table", click "User ID", "Access platform", "Statistical date", "Views", "number of visits", "Total jump" and "Total residence time", as shown in the following figure:

If the merge method is left merge and the merge basis is user ID, the access time, access times, residence time and other information of different users will be merged according to the Internet_ Match the "user ID" in the "user information table" and merge it into one table, as shown in the following figure:

Note: for details of "left and right merge", please refer to the file: "Basic functions of left and right merge".

Save the data table and update it.

2.3 View the effect

Get the access time, number of access, residence time and other information of different users according to the "Internet_User information table" after matching the "User ID", as shown in the following figure: