Unsupervised machine learning algorithms discover hidden patterns in unlabeled data without the need of human intervention.
Have you ever wondered how the evaluation of these algorithms take place with no ground truth labels?
Let’s find out 👇 (1/8)
@prthgo While doing a research internship, I came across this article that describes the current situation of XAI methods in the field of healthcare. https://t.co/ov3xurL1MA
There’s a lot of other ways to evaluate clusters as well. To learn more about Clustering and methods like silhouette analysis, you can visit my blog - https://t.co/Touw2Ow5Kj
This was my very first blog, so any kind of feedback is highly appreciated!
(8/8) 🍃🍃🍃
Unsupervised machine learning algorithms discover hidden patterns in unlabeled data without the need of human intervention.
Have you ever wondered how the evaluation of these algorithms take place with no ground truth labels?
Let’s find out 👇 (1/8)
Formula might look daunting, but we are just iterating through each data point calculating the distance between it and it’s assigned centroid. Squaring and summing of distance is a very common way to understand the variance.
This is the variance approach to evaluation
(7/8)
@prthgo Thanks a lot for the resource! Will check it out. The main reason I wanted to dig deep in XAI was that most of the new researches in the field of AI are black box systems. I feel like we are going down a dangerous path. What do you think?