Analyzing Data: Applying a Model

Once you've connected or uploaded your data sets, it's time to deploy one of our Smart AI™️ models. Each model uses different techniques to produce unique results based on the underlying data and the intent of the analysis. If you're unsure which type of analysis is most relevant to your current goal, you can see what type of questions each model answers by visiting the Data Settings tab (pictured below). 

Remember that each model is unique to the data stream it is connected to. Stratifyd allows you to tune each model to fit your analytic needs. With this flexibility, it is possible that data results can appear differently, depending on how you've tuned a model. 

How to Apply a Data Model

Each model deployment process is slightly different but generally follows the same steps.  Be sure to read the prompts in the platform to begin your deployment. Below, we've outlined the general steps. 

1. 1. You can apply a model from within a workspace, or you can add one to from Models tab. Here, we create it from within a workspace by selecting the workspace tab in the navigation pane.
   

    

   2. To access the Data Settings Panel menu, click the Data settings button, accompanied by the gear icon. 
   

   
   In the data settings panel that appears, you'll see a list of all data connections for that workspace.  Make sure you have selected the data stream you want to work with in the Connected column.
   
   

   

    

   3. In the Analyze tab, you'll want to select the drop-down that best describes your analytic goal. In this example I want to learn more about what my customers are saying.

   
   
   4.  Select the model that will satisfy the information required to perform your analysis. In this example, we want to identify the clusters of topics with bi-grams (two word combinations). This will give me high-level topic information for my entire data set. Click +Add to proceed.

 5.  On the next screen, you'll need to select the input fields required to generate results. The fields required depend on the model you have selected.  In this example, we want the topic clusters from our survey text. This will give us a good indication of what survey topics are occurring most in low and high survey scores. 


6.  Follow the remaining prompts and press Start Analysis to deploy the model. Then, hit Save in bottom right-hand corner to start model processing. 

Note: See model best practices for estimates on processing time required. 

Next Steps

Once you're ready to review your results, read this article for guidance on Visualizing your Results.

Further questions?

We're here to help! Don't hesitate to contact us for further assistance via chat or submit a ticket!