IBM Watson Projects
上QQ阅读APP看书,第一时间看更新

Refine

One of the more important IBM Watson tasks is Refine (which we mentioned earlier in this chapter). Here will walk through the basics, using some Watson sample data:

  1. From the Quick-start bar, click on Refine. From the Refine data set dialog, you then can scroll down and select Sample data; at this point you will see the Sample data dialog, which is displayed as follows:
  1. IBM Watson provides a nice list of sample data, each worth spending a bit of your time exploring and experimenting with. For now, let's pick Bike Sharing data set and then click on Upload:
To access the sample data it needs to be uploaded. Once uploaded it will appear as an informational tile (shown as follows) that can be selected and used:
  1. Now that we have our data loaded and available, you can select it from the Refine data set list, which automatically loads it into the Refine page (which looks a lot like an Excel worksheet):

There are many tasks you can perform using Refine, such as:

  • General housekeeping: Such as renaming columns, changing data types, or creating a subset of the data by filtering out irrelevant records
  • Summarization: By altering the default aggregations
  • Enrichment: By adding calculated fields, hierarchies and groups
  • Review the metrics of the data, such as a quality score by data field or column

For now, let's assume we've made some of the previously-mentioned refinements to our data and want to save it as a new file. To do that, you simply click on the SAVE icon (looks such as a tiny diskette in the upper left of the page), enter an appropriate name for the new file, and click Save (on the Save as popup shown as follows):

If you are working in a multi-user environment, your new (refined) dataset is saved by default in your personal folder. To share your refined dataset with others, move it to a shared folder.