Oracle Data Visualization Desktop (DVD) 4 New Features
In late October 2017, Oracle released updates to Data Visualization Desktop (DVD), Version 220.127.116.11 or “DVD 4,” a desktop solution that allows end users to build data visualizations without relying on their IT departments for assistance. The latest version adds several capabilities to the existing product which require modified installation, including a new home page; new data connector types; machine learning (ML) capabilities to make predictions and intelligent suggestions; data flow enhancements; data source editor changes; and updates to projects and visualizations including map backgrounds.
The installation of DVD 4 is very simple. The software is available on Oracle TechNetwork, in the Middleware section here. Extract the installer from the zip file. Double click the installer file “DVDesktop-18.104.22.168.0.exe,” choose the “Installation” folder, and click “Install”.
Once the installation is finished, notice the message on the final screen. It notes that Advanced Analytics needs to be installed to provide the correct version of R and related packages. Advanced Analytics can be installed by navigating to “Start > Programs > Oracle > Install Advanced Analytics.”
Optionally, install R and DVML. Selecting and running “Advanced Analytics” installs R. The install for R is much easier than in the past. Select the defaults to complete the R installation.
DVML is the Machine Language scripting tool (Python). If you plan to build machine learning scripts, install this component. DVML can be installed by navigating to “Start > Programs > Oracle > Install DVML.”
To avoid possible install issues, install DVML to a directory path that does not contain any spaces:
The new home page allows personalization and some helpful information for getting started with DVD. Click on the icon in the top left-hand side of the page to open the menu and to start a project or to prepare the data.
The home page menu provides access to the “Console.” The Console allows plug-ins import and viewing:
New data connectors are now available for Oracle Autonomous Data Warehouse Cloud (AWDC), Oracle Big Data Cloud Service and Oracle Talent Management Cloud. These new connectors; enhancements to the CSV data connector; the June enhancements to add ODBC, JDNC, OData and Oracle Docs; and the capability to source from Oracle Business Intelligence Enterprise Edition (OBIEE) subject areas and Oracle Fusion Applications with Oracle Transactional Business Intelligence mean that we now have a very wide range of data sources to access via DVD.
I wanted to mention one important pointer for the CSV data source. When uploading CSV and text files, you can override the auto-detected delimiter. Your choices are: “Comma, Semicolon, Space, Tab, and Custom.” You can also select “Custom,” and specify other delimiters.
Box plot, horizontal box plot, and waterfall visualizations were first introduced in DVD 3.
In addition, we can now add unrelated data sets to the same project and projects can contain visualizations from multiple, unrelated data sets. These data sets do not necessarily have to be joined to co-exist in the same project.
We can also now use background maps to enhance our geographical visualizations.
Data Source Editor
In previous DVD versions, the Data Source Dialog was used to create or edit data sources. The Data Source Dialog is replaced by the Data Source Editor, which provides the similar functionality but is more dynamic and easier to use.
With the “Data Flow” component, you can create your own data transformations, such as filtering and joining datasets; aggregation; and new columns based on custom formulas. Data flows can be stored and re-executed to bring in refreshed data. You can also apply custom scripts to data during the DVD transformation process. Finally, you can load data into an Essbase cube. Now, when you upload a file-based data source, columns containing numeric values are automatically imported as measures with the “Number” data type.
You can also take advantage of new binning features, including creating bins from measures, and using binning attributes to group data. You can apply a “Time Series Forecast” calculation to calculate additional rows of data by adding a “Time Series Forecast” to a data flow. You can create sequences and use them when building data sets; detect sentiment for a given text column by applying a sentiment analysis to a data flow; use cumulative aggregate functions to group your data; and merge rows from two data sets.
DVD 4 offers brand-new narration and storytelling features, which makes it easier to present data stories. Visualizations can be enhanced by adding links to related content under the “Data Actions” menu. You therefore have a wider range of number and date formats to choose the best display format for data in your visualizations. For quick and easy access, the properties of objects you select are now displayed in the “Data Panel.”
Many times, it is quicker to copy a visualization than starting from scratch. In DVD 4, we can copy and paste within the same canvas and between canvases in the same project. Using the “Duplicate” option, we can make copies of an object within the same canvas or duplicate the entire canvas.
In DVD 4, you can now use machine learning (ML) algorithms to show patterns and uncover insights in data sets, and then add them to visualizations. The “Machine Learning Catalog” is used to manage your ML scripts and models, and data flows are used to train ML models using custom or built-in scripts, as well as to score and predict data sets. You can create a custom scenario for attributes by applying ML models to a data set. DVD 4 also now offers a self-service ML model for diagnostic attribute analytics.