Our Predictive Anthropology platform sometimes identifies trends/topics that exhibit signs of volatility. These are topics where the topic universe (or cultural universe) moves in a haphazard manner. That is, the topics don't consistently converge towards a particular set of meanings but rather keep either diversifying completely or flip-flopping between greater and lesser diversity.

Let me give you an example.

Suppose you are searching for information on the topic of "nutrition" and you find that there are three or four major themes that cover the meanings of this concept. If these themes constantly change or are inconsistent, it can be difficult for consumers to understand the role that this trend or topic should play in their lives. As a result, they may try to apply the topic to different cultures and contexts. This indicates that the topic is volatile.

There are two types of volatility we measure.

The first is when the topic universe continues to diversify in a random pattern. In such cases, we do not run a prediction (it doesn't meet our quality criteria).

Instead, we report that the topic is highly volatile and can't be predicted.

The second is when the topic universe has recently begun to converge in a stable and predictable pattern, but high volatility in its history leaves us less certain about its continued trajectory.

In such cases, we run the prediction and report on the fact that the topic is newly stable.