Through different views, the content analytics miner user interface can show you trends and anomalies in your content set.
Many of these statistics are based on how frequently a particular facet value occurs. For example, you might see a rising number of references to a specific competitive product in your call center logs over time. For another example, you might notice frequent use of a particular acronym or word by certain people during an audit or investigation. This type of raw frequency analysis can be useful for quickly identifying hot terms or topics.
The content analytics miner, however, goes one step further to help you find high-frequency data that actually matters. In the preceding audit example, just because a term is used frequently by an individual does not mean that it is important. The real question is whether that term is used more frequently by that individual than by everyone else. With correlation ratings that are computed by the content analytics miner, you can quickly spot these types of anomalies, even if they are low-frequency.