Update README.md

README.md

@@ -47,13 +47,8 @@ size_categories:
 - 1M<n<10M
 ---
 
-[See a demo of a model trained on this dataset](https://huggingface.co/spaces/aivance/malphish-eater)
-
 **Important Notice:**
-- the data collection and analysis for `domain_registration_length` was incorrect
-- the functions to extract `external_favicons` and a few other features were not robust enough
-- 128,000 URLs in reserve were not utilized at all
-- therefore, a v2 is coming for the demo
+- A subset of the URL dataset is from Kaggle, and the Kaggle datasets contained 10%-15% mislabelled data. See [this dicussion I opened](https://www.kaggle.com/datasets/sid321axn/malicious-urls-dataset/discussion/431505) for some false positives. I have contacted Kaggle regarding their erroneous "Usability" score calculation for these unreliable datasets.
 
 <br>