Now that we've covered the architecture and flow, let's go over a number of concepts and terms that you should be aware of.
Unstructured (sometimes called textual) information is generally natural or free text. It is understood simply by people but is usually problematic for computer processing. In contrast, structured information is easily processed by a computer.
This is a term that refers to automated methods of converting unstructured data into structured data.
When you search, you already have something in mind that you are looking for, and your task is to create a search query that can locate an exact or partial match of your target.
Discovery refers to an exploratory exercise that is goal driven—this is what Watson refers to as starting points used to learn more about a specific topic.
Data mining is a method of locating patterns or insights within data. Data mining is a natural part of discovery.
An analytics collection is a grouping of documents that are indexed and available for search and analysis.
Facets represent the different characteristics of a collection and are used to navigate and analyze the collection with the miner. Default facets are created automatically for each collection. In addition, facets can be loaded with information obtained directly from structured data fields in your collection or with actual information from the text. Facets allow you to focus on only those documents that you are interested in.
Frequency represents the current number of documents that contribute to a selected keyword. This is useful in identifying trends in the collection.