Update README.md

README.md

@@ -1,3 +1,9 @@
----
-license: apache-2.0
----
+Models used in training for parts of OntoUML Image Taxonomy Extractor (OITE)
+The second dataset consists of the 643 OntoUML images[source](https://doi.org/10.1007/978-3-031-17995-2_1), 643 randomly picked class diagrams and 643 randomly picked non-UML images, taken from the dataset of the paper “Multiclass classification of four types of UML diagrams from images using deep learning" [link](https://doi.org/10.5281/zenodo.4595957 
+). In my paper this dataset is referred to as the 3-class dataset. 
+See: https://github.com/SimeonKaishev/OntoUML_IMG_Converter
+
+
+---
+license: apache-2.0
+---