Category: Big Data & Analytics

Tableau – Filter Operations ”; Previous Next Any data analysis and visualization work involves the use of extensive filtering of data. Tableau has a very wide variety of filtering options to address these needs. There are many inbuilt functions for applying filters on the records using both dimensions and measures. The filter option for measures offers numeric calculations and comparison. The filter option for dimension offers choosing string values from a list or using a custom list of values. In this chapter, you will learn about the various options as well as the steps to edit and clear the filters. Creating Filters Filters are created by dragging the required field to the Filters shelf located above the Marks card. Create a horizontal bar chart by dragging the measure sales to the Columns shelf and the dimension Sub-Category to the Rows shelf. Again drag the measure sales into the Filters shelf. Once this filter is created, right-click and choose the edit filter option from the pop-up menu. Creating Filters for Measures Measures are numeric fields. So, the filter options for such fields involve choosing values. Tableau offers the following types of filters for measures. Range of Values − Specifies the minimum and maximum values of the range to include in the view. At Least − Includes all values that are greater than or equal to a specified minimum value. At Most − Includes all values that are less than or equal to a specified maximum value. Special − Helps you filter on Null values. Include only Null values, Non-null values, or All Values. Following worksheet shows these options. Creating Filters for Dimensions Dimensions are descriptive fields having values which are strings. Tableau offers the following types of filters for dimensions. General Filter − allows to select specific values from a list. Wildcard Filter − allows to mention wildcards like cha* to filter all string values starting with cha. Condition Filter − applies conditions such as sum of sales. Top Filter − chooses the records representing a range of top values. Following worksheet shows these options. Clearing Filters Filters can be easily removed by choosing the clear filter option as shown in the following screenshot. Print Page Previous Next Advertisements ”;

Tableau – Top Filters ”; Previous Next The Top option in Tableau filter is used to limit the result set from a filter. For example, from a large set of records on sales you want only the top 10 values. You can apply this filter using the inbuilt options for limiting the records in many ways or by creating a formula. In this chapter, you will explore the inbuilt options. Creating a Top Filter Using the Sample-superstore, find the sub-category of products which represents the top 5 sales amount. To achieve this objective, following are the steps. Step 1 − Drag the dimension Sub-Category to the Rows shelf and the Measure Sales to the Columns shelf. Choose the horizontal bar as the chart type. Tableau shows the following chart. Step 2 − Right-click on the field Sub-Category and go to the tab named Top. Here, choose the second radio option by field. From the drop-down, choose the option Top 5 by Sum of Sales. On completion of the above step, you will get the following chart, which shows the top 5 Sub-Category of products by sales. Print Page Previous Next Advertisements ”;

Tableau – Extracting Data ”; Previous Next Data extraction in Tableau creates a subset of data from the data source. This is useful in increasing the performance by applying filters. It also helps in applying some features of Tableau to data which may not be available in the data source like finding the distinct values in the data. However, the data extract feature is most frequently used for creating an extract to be stored in the local drive for offline access by Tableau. Creating an Extract Extraction of data is done by following the menu – Data → Extract Data. It creates many options such as applying limits to how many rows to be extracted and whether to aggregate data for dimensions. The following screen shows the Extract Data option. Applying Extract Filters To extract a subset of data from the data source, you can create filters which will return only the relevant rows. Let’s consider the Sample Superstore data set and create an extract. In the filter option, choose Select from list and tick mark the checkbox value for which you need to pull the data from the source. Adding New Data to Extract To add more data for an already created extract, you can choose the option Data → Extract → Append Data from File. In this case, browse the file containing the data and click OK to finish. Of course, the number and datatype of columns in the file should be in sync with the existing data. Extract History You can verify the history of data extracts to be sure about how many times the extract has happened and at what times. For this, you can use the menu – Data → Extract History. Print Page Previous Next Advertisements ”;

Tableau – Date Calculations ”; Previous Next Dates are one of the key fields which is extensively used in most of the data analysis scenarios. Hence, Tableau provides a large number of inbuilt functions involving dates. You can carry out simple date manipulations such as adding or subtracting days from a date. You can also create complex expressions involving dates. Following are the steps to create a calculation field and use date functions in it. Create Calculated Field While connected to Sample superstore, go to the Analysis menu and click ‘Create Calculated Field’, as shown in the following screenshot. Calculation Editor The above step opens a calculation editor, which lists all the functions available in Tableau. You can change the dropdown value and see only the functions related to Date. Create a Formula Now, find out the sales volume along with the difference in the date of sales in months from 21st March 2009. For this, create the formula as shown in the following screenshot. Using the Calculated Field Now to see the created field in action, you can drag it to the Rows shelf and drag the Sales field to the Columns shelf. Also drag the ship Date with months. The following screenshot shows the Sales values. Print Page Previous Next Advertisements ”;

Tableau – Data Blending ”; Previous Next Data Blending is a very powerful feature in Tableau. It is used when there is related data in multiple data sources, which you want to analyze together in a single view. As an example, consider the Sales data is present in a relational database and Sales Target data in an Excel spreadsheet. Now, to compare actual sales to target sales, you can blend the data based on common dimensions to get access to the Sales Target measure. The two sources involved in data blending are referred as primary and secondary data sources. A left join is created between the primary data source and the secondary data source with all the data rows from primary and matching data rows from secondary data source. Preparing Data for Blending Tableau has two inbuilt data sources named Sample-superstore and Sample coffee chain.mdb which will be used to illustrate data blending. First load the sample coffee chain to Tableau and look at its metadata. Go to the menu – Data → New Data Source and browse for the sample coffee chain file, which is a MS Access database file. The following screenshot shows the different tables and joins available in the file. Adding Secondary Data Source Next, add the secondary data source named Sample-superstore by again following the steps – Data → New Data Source and choosing this data source. Both the data sources now appear on the Data window as shown in the following screenshot. Blending the Data Now you can integrate the data from both the above sources based on a common dimension. Note that a small chain image appears next to the dimension named State. This indicates the common dimension between the two data sources. Drag the State field from the primary data source to the rows shelf and the field Profit Ratio from the secondary data source to the Columns shelf. Then, select the bullet chart option from Show Me to get the bullet chart shown in the following screenshot. The chart shows how the profit ratio varies for each state in both the superstore and coffee chain shops. Print Page Previous Next Advertisements ”;

Tableau – Custom Data View ”; Previous Next A custom data view is used to extend the normal data views with some additional features so that the view can give different types of charts for the same underlying data. For example, you can drill down a dimension field which is part of a pre-defined hierarchy so that additional values of the measures are obtained at a different granularity. Following are some of the frequently used and important custom data views Tableau offers. Drill Down View For dimension fields which are part of a hierarchy, you usually need to know the result of analysis for the next or previous level of aggregation. For example, when you know the result for a quarter, you get interested to know the results for each month in that quarter, and you may even need the result for each week. This is a case of drilling down the existing dimensions to get a finer level of granularity. To drill down and drill up for individual dimension members in a hierarchy, right-click a table header and select Drill Down from the context menu. Consider a bar chart created with the dimension category in the columns shelf and the measure Sales in the Rows shelf. Right-click on the bar representing Furniture and select Drill Down. The result of the drill down action is shown in the following screenshot. Swapping Dimensions You can create a new view from an existing view by swapping the position of the dimensions. This does not change the values of the measures, but it does change the position of the measures. Consider a view for analyzing the Profit for each year for each segment and category of products. You can click on the vertical line at the end of category column and drag it to the segment column. This action is shown in the following screenshot. The result of the swapping of the two dimensions is shown in the following screenshot. As you can see, only the position of the values of the measure Profit changes for each category and segment, and not its value. Print Page Previous Next Advertisements ”;

Tableau – Data Joining ”; Previous Next Data joining is a very common requirement in any data analysis. You may need to join data from multiple sources or join data from different tables in a single source. Tableau provides the feature to join the table by using the data pane available under Edit Data Source in the Data menu. Creating a Join Consider the data source ‘Sample superstore’ to create a join between Orders and Returns table. For this, go to the Data menu and choose the option Edit Data Source. Next, drag the two tables, Orders and Returns to the data pane. Depending on the field name and datatype, Tableau will automatically create a join which can be changed later. The following screenshot shows the creation of an inner join between Orders and Returns using the Field Order ID. Editing a Join Type The type of join which the table creates automatically can be changed manually. For this, click the middle of the two circles showing the join. A popup window appears below which shows the four types of joins available. Also Tableau automatically greys out some types of joins, which it finds irrelevant on the basis of data present in the data source. In the following screenshot, you can see the inner and left outer join as the available joins. Editing Join Fields You can also change the fields forming the join condition by clicking the Data Source option available in the join popup window. While selecting the field, you can also search for the field you are looking for using a search text box. Print Page Previous Next Advertisements ”;