Dependency parsing is a way to understand the relationships between words in a sentence. Dependency relations are more fine-grained attributes available to help build the model's understanding of the words through their relationships in a sentence. 

doc = nlp(sentenses[2])
spacy.displacy.render(doc,style='dep', options={'distance' : 140}, jupyter=True)

 These relationships between words can get complicated, depending on how sentences are structured. The result of dependency parsing a sentence is a tree data structure, with the verb as the root as shown in the figure below.