The short answer is you don’t get a positive or negative indication simply from our engine.
This is done by design. Sentiment is a fallacy in the world of big data because it is completely subjective, and there is no way (at least not yet) for a machine to truly understand it.
Consider the following example. You search for whole grains and find that one of the related topics is high fat. This could be a positive thing if it refers to high in healthy fats, or a negative thing if it refers to high fats/putting on weight. Depending on the level of knowledge of the person engaging on the topic, it's highly likely that both meanings exist.
The point isn't whether it's positive or negative. The point is that high fat is associated with whole grains. And we have to determine if high fat offers a viable strategy to pursue in relation to whole grain.
There are, of course, some shortcuts we can use to assess the value of high fat in the context of whole grains. One of which is to add high fat as an additional topic into the search bar and then examine the related articles. This will provide further insight into the intersection point between the two topics (whole grain and high fat) and show us the dominant associations made between the two topics in culture.
