How can find the shortest path length between two words by using WordNet?

The easiest way is to translate the Wordnet structure into a graph. Let's define synsets as nodes and relations as edges. Obviously, there might be several edges between the same nodes, but they will be labeled. You can assign each label a cost based, e.g., on the similarity factor from Chapter Estimating Semantics Distance of Texts Based on Used Terms Analysis

Having that you can simply implement any of the shortest paths algorithms.