For this exercise our goal is to get you on the path of facilitating innovation based on a model you've already created. If you're not familiar with creating a model then do this model building exercise first, then return to this one.
In this exercise you will be building an Excel spreadsheet that contains information about the elements in your model, and we will be use the Cognitive City's bulk import tool to load that data into your model. You'll then be able to create a network view that shows how the information from the spreadsheet connects together within the context of your model. Here are the steps:
Step 1: Review the model you created in your Cognitive City. This model should have included an element type that represented people you might invite to a facilitated conversation and artifacts that would be relevant to them. Now we need some actual data. Collect or create a spreadsheet about your people and your artifacts. This dataset must meet the following criteria for this assignment:
- It must include 10-100 artifacts and at least 5 people.
- It must include some data about both the people and the artifacts.
- The data for the people should include a column that contains a list of multiple values. For example, “expertise” or “datasets worked with” or “countries lived in”, etc.
- The data for the artifacts should include a column that contains unstructured text of at least a few sentences. For example, “description”, “abstract”, “definition”, “explanation”, “notes”, etc.
If you can’t collect or create a dataset that meets these criteria, then you should change your selection of people or artifacts until you can.
Step 2: As you collect or create your dataset, don't worry if you think of ways you should change your model. Part of having a "graph mindset" is not being afraid to evolve your data model as your thinking evolves. This is exactly what graphs are made for. Feel free to modify your existing model, or even build a new model, that can co-exist next to your old one, to represent your dataset. Also don't worry if your model contains many more elements than what you have data for. It's okay to import data into just a subset of your model. Because your dataset should meet the criteria outlined in step 1, it should be pretty simple to build a model for it: a starting pair that connects the people to the artifacts and then one node connected to the people that holds the additional structured information about them and one node connected to the artifacts to hold additional structured information about the artifacts. For example, a model that is based on TED Talks could look like:
You shouldn’t have to change the fields in the model all that much except for the artifacts in order to add a field to hold the unstructured text data and any other properties unique to each artifact. For example, on the TED Talks model above, I renamed the default “Name” property to “Title” and added a long text “Description” as well as an image and url fields. I also renamed the auto-generated “Tag” field to “Tags” since I knew that every talk would have multiple tags linked to it.
Step 3: Import your data by getting an Excel workbook of data based on your model, import it, and create some simple views to make sure it imported as expected. Once you’ve created your model, it should be straightforward to import data as long as your spreadsheet is in the following format:
- The sheet names match the names of the element types you are importing.
- The column names match the names of the fields for each element type.
If you want to see the correct format, click on “Data” in the left Admin menu bar and then click on the “Download Blank Example” button. Once you have data in a spreadsheet of this format, drag it back into the bulk import tool to upload. You can review this helpdesk article for more details about importing data.
Step 4: Once you’ve imported your data, create your first network view! Experiment with using the view to make sure that any fields that came from lists of multiple values in your spreadsheet got correctly split into separate nodes. If you aren’t seeing these split up, make sure you set the delimiter correctly during the data import. You may need to delete the data (use the legend in the view to select all elements of a particular type, and then click the delete button in the top of the right side panel.) and re-import.
That's it for this exercise. In the previous exercise you created your first model, now you've created your first view! This is the foundation of information assembly. In the next exercise we will see how to assess the network of information you've assembled (which might at first look like a tangled mess or a useless set of disconnected nodes) for it's innovation potential.