I have developed a new technique. Although I have performed multiple experiments and obtained various "meaningful" results, I am unsure about how to evaluate the technique's performance and ensure the confidence of the obtained results since there are no other techniques available for comparison. What are the best practices for evaluating and validating a new technique in the absence of a benchmark tool or dataset?
I think your own results should explain the value of your method, those reached the moon didn't compare their method to any other one!! Wish you luck.
Thank you. I feel like there's a bit of a chicken-and-egg situation here. I want to confirm that my results are valid, but to do so, I need to evaluate them. At the same time, I'm hoping that my results will provide insight into the value of my research.
It's important to remember that results do not necessarily speak for themselves. In other words, the results alone may not necessarily explain their value.
Dear Hussien Al-haj Ahmad
There can be a set of characteristics for each technology and, as a result, each can be measured against standards. In other words, the objectives of this technology are specific, and each original objective can be divided into a set of sub-objectives, so that more layers are achieved to specific standards related to those in any of the fields or industries, and as a result, they can be generalized. It is also possible to compare with the current state to verify the validity of the assumptions and the practical applicability of them, and also to benefit from experts in the field to obtain points of view and infer from them, for example, it's possible to benefit from
(National Institute of Standards and Technology)
I would argue you can always compare to some state-of-the-art techniques. Then you can build up the comparison as "To evaluate our technique, we compare it to . Although works inherently different than our technique, comparing results enables us to evaluate how we measure up to state-of-the-art."
In the discussion part you can then discuss the results of your technique and how it relates to state-of-the-art. Is your performance better? Do you make a better trade-off between criteria? Are you approximately as good as state-of-the-art but easier to apply? etc.
As long as you explain why you set up a certain comparison and why you select specific state-of-the-art techniques, the evaluation has value.
Its best to use the pre-experimental design it does not have a comparison group. The one-shot case study compares the post-test results to the expected results.
When you use the term technique shall I assume it a variation on a method? I wonder then if your method is so different that it cannot be validated by a standard process available. So, the questions seems to be whether you are validating the method or the results?
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts