Pentaho Analyzer is an analytical visualization tool that filters and drills into data sources. It can easily compile data into visualizations using an interactive web interface and is intuitive for end users. This plugin is available in the Pentaho Enterprise Edition only. If you are using the MongoDB native connection, it is possible to create intuitive and quick visualizations.
This recipe will guide you through exploring the Foodmart data imported in the previous recipe using this analytical tool. This recipe is an overview of the main features available in Pentaho Analyzer.
To get ready for this recipe, you will need the Pentaho BI server running, and make sure you have the MongoDB server running with the data from the previous recipe.
After you have made sure that you are ready to start the recipe, perform the following steps to create an analysis report:
- From the web browser, access the Pentaho BI server using
http://localhost:8080/pentaho. - After logging in, on the Home screen, click on the Create New button and then on the Analysis Report button, as you can see in the following screenshot:
- In the Select Data Source popup, select mongoFoodmart: Sales, as you can see in this screenshot:
- You'll get an interactive report with fields in the left-hand-side bar, as you can see in the following screenshot. You can drag and drop, using your mouse, those fields in Rows, Columns, or indeed Measures, if that be the case.
- Let's answer some business questions, such as What is my profit by country and state? You just need to drag and drop the Country and State Province fields from the Customers group to the Rows area. You also need to drag and drop the Profit field from the Measures group to the Measures area, and you will instantly get the answer to your business question, as you can see in this screenshot with the result:
- Now that we have got the first answer, we can keep asking more questions, such as "What is my profit by country and state for different product families?" Using the same view as earlier, drag and drop the Product Family field from the Products group to the Columns area, and you will quickly get your answer, as shown in the following screenshot:
- Let's answer one more question, What is my profit by in the cities of California for different product families? In keeping with the previous view, right-click on the
CAstate and choose Keep Only CA. Then, drag and drop the City field from the Customers group to the Rows area and you will get your answer, as you can see in the following screenshot: - As there are many cities in California, it isn't possible to get any quick understanding of the previous result. Let's convert the table into a stacked column chart in order to get a better understanding. To do this, click on the chart icon in View As and choose Stacked Column. You will get a quick visualization of your question, as you can see in the next screenshot:
- Let's save this view by clicking on the Save button in the top bar. In the Save popup, enter the name
California profitand save it in your user folder using theadminuser, as you can see in the following screenshot:
With a database and an OLAP schema on top, Pentaho Analyzer can answer multiple business questions quickly and with an intuitive user interface. In this recipe, we just perform some analyses of the Foodmart database, getting quick results in a dynamic table that can be swapped with a proper chart to get a better understanding of the data. This intuitive interface is possible with the OLAP schema, which gives the necessary metadata to the end user. The interface will execute MDX statements onto the Mondrian engine, and then Mondrian will translate the statements on the fly into MongoDB using aggregation pipeline queries.
There are more features available in Pentaho Analyzer; for example, it is possible get the totals by rows or measure columns as a subtotal breaking by hierarchy.
These analysis views can easily be embedded in other systems or plugins; for example, in the next recipe, we will embed this view in a Pentaho dashboard.
In the Pentaho marketplace, there is an open source plugin similar to Pentaho Analyzer named Saiku Analytics. It's a lightweight JavaScript web interface, and a backend based on web services makes it easier to embed. Saiku also has an enterprise version, just like Pentaho, with extra features that aren't available in the community edition.