Update README.md

README.md

@@ -7,4 +7,13 @@ language:
 - en
 pipeline_tag: text-classification
 ---
-This model classifies sentiment of the scientific text based on it's context, i.e text from scientific journals to negative (n), positive (p) and neutrals (o).
+This model is finetuned on Scibert. 
+It classifies sentiment of the scientific text based on it's context.
+Here, the context is the sentiment of the author depicted in the sentence. 
+There are three classes; postitive, negative and neutral.
+This model outputs following classnames according to the sentiment:
+p = positive
+n = negative
+o = other/neutral
+
+For finetuning, the publicly available dataset on context identification from Angrosh et al. https://dl.acm.org/doi/10.1145/1816123.1816168 is used.