In [20]:

    

import py2neo
import pandas as pd

graph = py2neo.Graph()

data = graph.data("MATCH (f:Field)<-[:DECLARES]-(c:Class)-[:DECLARES]->(m:Method), (m)-[r:READS|WRITES]->(f) RETURN m.name, f.name, r.lineNumber")
df = pd.DataFrame(data)
df['hit'] = 1
df.head()


    

    
Out[20]:







  

    

      

      
f.name
      
m.name
      
r.lineNumber
      
hit
    
  
  

    

      
0
      
duration
      
getPrincipalFee
      
49
      
1
    
    

      
1
      
duration
      
extend
      
24
      
1
    
    

      
2
      
duration
      
extend
      
24
      
1
    
    

      
3
      
duration
      
<init>
      
18
      
1
    
    

      
4
      
dailyRate
      
getPrincipalFee
      
49
      
1
    
  

In [21]:

    

matrix = df.drop_duplicates().pivot(index="m.name", columns="f.name", values="hit").fillna(0)
matrix


    

    
Out[21]:







  

    

      
f.name
      
customer
      
dailyRate
      
date
      
duration
      
fees
    
    

      
m.name
      

      

      

      

      

    
  
  

    

      
<init>
      
1.0
      
1.0
      
1.0
      
1.0
      
1.0
    
    

      
addFee
      
0.0
      
0.0
      
0.0
      
0.0
      
1.0
    
    

      
extend
      
0.0
      
0.0
      
0.0
      
1.0
      
0.0
    
    

      
extendForWeek
      
1.0
      
1.0
      
1.0
      
0.0
      
0.0
    
    

      
getAdditionalFees
      
0.0
      
0.0
      
0.0
      
0.0
      
1.0
    
    

      
getPrincipalFee
      
1.0
      
1.0
      
0.0
      
1.0
      
0.0
    
  

In [22]:

    

from sklearn.metrics.pairwise import cosine_distances

dissimilarity_matrix = cosine_distances(matrix)
dissimilarity_matrix[:5,:5]


    

    
Out[22]:





array([[ 0.        ,  0.5527864 ,  0.5527864 ,  0.22540333,  0.5527864 ],
       [ 0.5527864 ,  0.        ,  1.        ,  1.        ,  0.        ],
       [ 0.5527864 ,  1.        ,  0.        ,  1.        ,  1.        ],
       [ 0.22540333,  1.        ,  1.        ,  0.        ,  1.        ],
       [ 0.5527864 ,  0.        ,  1.        ,  1.        ,  0.        ]])

In [23]:

    

from sklearn.manifold import MDS

# uses a fixed seed for random_state for reproducibility
model = MDS(dissimilarity='precomputed', random_state=0)
dissimilarity_2d = model.fit_transform(dissimilarity_matrix)
dissimilarity_2d[:5]


    

    
Out[23]:





array([[ 0.01921689,  0.12978066],
       [ 0.3465834 , -0.43509102],
       [-0.5699487 , -0.06500858],
       [ 0.09208679,  0.49470377],
       [ 0.3478405 , -0.43414008]])

In [24]:

    

%matplotlib inline
import matplotlib.pyplot as plt

plt.figure(figsize=(8,8))
x = dissimilarity_2d[:,0]
y = dissimilarity_2d[:,1]
plt.scatter(x, y);