Data clusters are visualizations that identify values that are likely to be related. In this video, learn how to identify data clusters.
- [Instructor] When you visualize your data using a scatterplot, you'll often find that some of the values form obvious groups. In Machine Learning and Data Analysis terms, those groups are called clusters. One of the great features in Tableau 2020 is the ability to identify those clusters within your data. In this movie, I will show you how to do that. My sample file is 0905 clusters, and you can find it in the chapter nine folder of the exercise files collection. In this workbook, I have a worksheet that map's an individual's income versus their sales total. If we take a look at the data source by going down to the bottom left corner and clicking data source, you can see that I have income in thousands, and that is mapped to their sales total in actual dollars. So, the first individual has an income, household income of 85000 sales total of 225, 133000 lead to sales of 1128, and so on. So the idea is that I would like to identify different types of customers. Those who have high income and spend a lot of money, those who have low income and spend a lot of money, and so on. So I will go back to sheet one and from here, we can create our cluster. I'll go to the Analytics tab over in the navigation pane and drag cluster on to the visualization. And you can see here that there is a cluster button as a target. So I'll drop there and you can see that the program has identified three clusters and I have the clusters dialog box here. The variables it's using are Income and Sales Total, which is good, those are the only two I have and the number of clusters is automatic. So I'll close this box here and you can see that I have three clusters of customers. Income is on the x or horizontal axis, and Sales Total is on the vertical axis. So it looks like that customers with higher incomes do in fact, tend to spend more but I've got a lot of customers in the middle who spend a fair amount as well. So that's a good sign. You're not limited to the clusters that Tableau creates for you though. You can change the number of clusters for example. To do that, I will go to the Clusters pill, which is on the marks card, click this down arrow, and I will click Edit clusters. This is the dialog box we saw earlier and here you have the number of clusters box currently is set to automatic, but if I want to I can define four clusters. So I'll type four and the analysis is updated in the background so I'll close the dialog box and we can see that it looks like cluster number two which is the higher income has stayed constant but there are subdivisions within what used to be cluster number two. I can see more details about my clusters by going to the Clusters pill, again this on the marks card, click the down arrow, and click Describe clusters. On the Summary tab of the Describe clusters dialog box, I have information about the number of clusters, the number of points and various values that are statistical descriptions of what's going on in the analysis. Down below, there is a table with the number of items in each cluster, and also other values of their centers. In other words, if you were to take the average position of all the points within a cluster, that's what sum of Income and sum of Sales Total would be for each of the clusters. And it turns out that no points fell outside of this analysis so everything is part of a cluster. If you want to learn more about these statistics, you can go down and click the learn more about the cluster summary statistics link at the bottom of the dialog box. On the Models tab, we have statistical analysis of variance or ANOVA and there we have our P values, which is a value indicating goodness of fit. Anything below point 01 is considered a very good fit and you'll see that the values here are extremely low, much lower than point 01. When you're done looking at the analysis, you can click the Close button and if you want to remove the clustering, all you need to do is drag clusters from the marks card off of the visualization, and you're back to the scatterplot that you started with.
- Explain where a user would navigate to seek specific help in Tableau.
- Determine the best approach for using Excel in Tableau.
- Interpret how to use the features and functions of Tableau when creating charts.
- Describe how best to manage data in a worksheet or visualization.
- Explain how to create a selection filter for certain values.
- Explain how to manage data for different chart formats.
Skill Level Beginner
1. Introducing Tableau
2. Managing Data Sources and Visualizations
3. Managing Tableau Worksheets and Workbooks
4. Creating Custom Calculations and Fields
5. Analyzing Data
6. Sorting and Filtering Tableau Data
7. Defining Groups and Sets
8. Creating and Pivoting Crosstabs
9. Creating Basic Visualizations
10. Formatting Tableau Visualizations
11. Annotating and Formatting Visualizations
12. Mapping Geographic Data
13. Creating Dashboards and Actions
Next steps1m 17s
- Mark as unwatched
- Mark all as unwatched
Are you sure you want to mark all the videos in this course as unwatched?
This will not affect your course history, your reports, or your certificates of completion for this course.Cancel
Take notes with your new membership!
Type in the entry box, then click Enter to save your note.
1:30Press on any video thumbnail to jump immediately to the timecode shown.
Notes are saved with you account but can also be exported as plain text, MS Word, PDF, Google Doc, or Evernote.