Go to content

The confidence index revisited

Plant classification according to a revised confidence index

The confidence index revisited

The confidence index tells us which plants to focus on according to the guidelines of plant authors.

When several plants are proposed for a given symptom, classifying them in one way or another seems to be the way to go. In a previous post, we highlighted a confidence index.

The confidence index is not really a ranking in order of importance, that would be a bit too presumptuous on our part, and in any case, if the plant comes out at the time of the research it is because one of the phytotherapy specialists suggested it as appropriate for the intended use.

Rather, the confidence index is a way of showing the degree of unanimity among specialists regarding the traditional uses of plants.

While the aim of the project is to maintain objectivity over time (in fact, we do not give our opinion on plants), we simply gather information from authors specialized in their field, and we regurgitate it concisely. Just for the display, we wanted to do a little bit better than the alphabetical order.

Our first calculation was worthy of a kindergarten student, it was almost indecent. The example below illustrates the simplicity of our initial thought.

  • Out of 9 books, 9 recommend yarrow for hemorrhoids
  • Out of 9 books, only 1 recommends wild garlic for hemorrhoids
  • On a scale of 5, it gave us:
  • 9 / 9 = 1 x 5 = 5
  • 1 / 9 = 0.11 x 5 = 0.55 rounded up to the next unit = 1
  • To conclude, we got:
  • a confidence index of 5 out of 5 (🟩 🟩 🟩 🟩 🟩) for yarrow
  • a confidence index of 1 out of 5 (🟩 ⬜ ⬜ ⬜ ⬜) for wild garlic

Having successfully completed our transition to preschool, we have refined our method somewhat. Now we take a few more parameters into account. Let us consider, at random, the case of diarrhea.

Example 1: To know the burr index for diarrhea, we need:

  • A=13 (sources that mention bramble as appropriate for diarrhea)
  • B=18 (sources mentioning other plants as suitable for diarrhea)
  • C=16 (sources that mention all-purpose bramble)
  • T=20 (total number of sources)
  • A/B (reminder) = 0.72 (72%)
  • A/C (precision) = 0.81 (81%)
  • β (coefficient based on all sources) = square root of C/T = 0.89
  • weighted average = square root of (A/B) * (A/C) * β = 0.69 (69%)
  • index = 4 out of 5

Example 2: to know the index of the spotted deadnettle for diarrhea, we need:

  • A=1 (sources that mention spotted deadnettle as appropriate for diarrhea)
  • B=18 (sources mentioning other plants as suitable for diarrhea)
  • C=1 (sources that mention the all-purpose spotted deadnettle)
  • T=20 (total number of sources)
  • A/B (reminder) = 0.06 (6%)
  • A/C (precision) = 1.00 (100%)
  • β (coefficient based on all sources) = square root of C/T = 0.22
  • weighted average = square root of (A/B) * (A/C) * β = 0.05 (5%)
  • index = 1 out of 5

In conclusion, the calculation of the index, even though it now looks a bit like a gas factory, seems more accurate, especially in cases where there are poor representations of plants in books. A high index is therefore avoided for a use that has only a few sources. And finally, in view of the development of our mathematical knowledge, we can’t wait to go to primary school!

If you are interested in our search tool, more info at phyto-info.com.

Discover our Plant Search Tool

And if you have any questions, please do not hesitate to contact us 🤗

Stay tuned !

Back to top