Added description of the methods

README.md

@@ -17,9 +17,7 @@ We develop two approaches for norm conflict identification, *i.e.*, classify uns
 Our approach to identify norm conflicts are based on generating comparative embeddings by computing the distance between norm embeddings *Emb<sub>norm</sub>* and using these distances as a semantic representation of the presence or absence of norm conflicts (*i.e.*, conflicts and non-conflicts). We developed two such distance-based approaches to find the comparative embedding corresponding to the conflicts between norms, both of which are based on finding the mean value (centroid in embedding space). The first approach consists of identifying the centroid of concatenated embeddings of norm pairs representing conflicts and the centroid of concatenated embeddings representing non-conflicts. The centroid is generated by computing the mean of this extended embedding space, which we call *Emb<sub>cent</sub>*. The second approach consists of computing two centroids for the offset embeddings of norm pairs. 
 The centroid is generated by computing in the same embedding space as the original norms 
 
-```math
-Emb_cent = \frac{\sum_{i=1}^{C}Emb_{off} {[i]}}{C}
-```
+$Emb_{cent} = \frac{\sum_{i=1}^{C} Emb_{off}{[i]}{C}$
 
 where