SPSS vs Azure Machine Learning (ML) in Data Prep and Storage

Posted By:

One of the major differences between SPSS and Azure ML lies in the way datasets are managed and stored.  Since Azure ML stores datasets on the cloud, it allows the user to utilize and manipulate multiple datasets at once.  The user has access to all of their datasets in one place, and Azure ML allows the user to drag and drop any of the datasets they wish to use on the workspace.  For example, the user can combine datasets or take columns of data from one dataset and add it to another in a very simple way.

Another difference between SPSS and Azure ML is the interface.  Azure ML is more visual and allows the user to see how datasets are connected and/or manipulated on the workspace.  The data can be viewed at any point throughout the data cleaning/manipulation process.  For example, when filtering out multiple data points in SPSS, separate datafiles will likely need to be created that each contain segments of data.  If the user needs to analyze data from any of the previous data splits, they will have to open the previous dataset to begin analyzing.  It can be time consuming and confusing the work with multiple datasets in that way. With Azure ML, all of the data splits and filters are on one workspace, which allows the user to view and manipulate these datasets at any point in the splitting/filtering process.   Azure ML provides a much simpler, and more intuitive way to manage and prep data.