

In [1]:

    
import sys

sys.path.append('../../code/')
import os
import json
from datetime import datetime
import time

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import scipy.stats as stats

import networkx as nx

from load_data import load_citation_network, case_info
from helper_functions import *

%load_ext autoreload
%autoreload 2
%matplotlib inline

data_dir = '../../data/'
court_name = 'scotus'



In [22]:

    
court_adj_mat = pd.read_csv(data_dir + 'clean/jurisdictions_adj_mat.csv', index_col='Unnamed: 0')
court_adj_mat.index = [j + '_ing' for j in court_adj_mat.index]
court_adj_mat.columns= [j + '_ed' for j in court_adj_mat.columns]

Grab federal appellate subnetwork



In [44]:

    
fed_appellate = ['ca' + str(i+1) for i in range(11)]
fed_appellate.append('cafc')
fed_appellate.append('cadc')

fed_appellate_ing = [j + '_ing' for j in fed_appellate]
fed_appellate_ed = [j + '_ed' for j in fed_appellate]



In [45]:

    
fed_appellate_network = court_adj_mat.loc[fed_appellate_ing, fed_appellate_ed]



In [46]:

    
fed_appellate_network

get inter court favorability rankings



In [73]:

    
inter_juris_rankings = pd.DataFrame(columns = fed_appellate)

for court in fed_appellate:

    # grab the inter court citations for court
    citing_count = fed_appellate_network.loc[court + '_ing']
    
    # rank the courts by number of citations
    court_ranking = citing_count.sort_values(inplace=False, ascending=False).index.tolist()
    inter_juris_rankings[court] = [j.split('_')[0] for j in court_ranking]
    
inter_juris_rankings.index = [i + 1 for i in inter_juris_rankings.index]
inter_juris_rankings.index.name = 'rank'



In [74]:

    
inter_juris_rankings

load jurisdiction network into igraph



In [75]:

    
import igraph as ig



In [80]:

    
court_adj_mat.as_matrix() + court_adj_mat.transpose().as_matrix()









    Out[80]:





array([[500928,  95606,  84758, ...,     56,      1,      0],
       [ 95606, 303404,  10885, ...,     25,      0,      0],
       [ 84758,  10885, 186844, ...,     28,      0,      0],
       ..., 
       [    56,     25,     28, ...,   3062,      0,      0],
       [     1,      0,      0, ...,      0,      0,      0],
       [     0,      0,      0, ...,      0,      0,      0]])



In [92]:

    
num_jurisdictions = court_adj_mat.shape[0]

courts = [j.split('_')[0] for j in court_adj_mat.index.tolist()]



In [93]:



In [116]:



In [131]:

    
G = ig.Graph(n=num_jurisdictions)
G.vs['name'] = courts

# list of all edges
edges = get_pairs(jurisdictions)
for e in edges:
    ct1 = e[0]
    ct2 = e[1]

    weight = court_adj_mat.loc[ct1 + '_ing', ct2 + '_ed'] + court_adj_mat.loc[ct2 + '_ing', ct1 + '_ed']
    
    if weight > 0:
        v1 = G.vs.select(name=ct1)[0]
        v2 = G.vs.select(name=ct2)[0]
        
        G.add_edge(v1, v2, weight=weight)



In [133]:

    
G.summary()









    Out[133]:





'IGRAPH UNW- 412 41261 -- \n+ attr: name (v), weight (e)'



In [146]:

    
visual_style = {}
visual_style["layout"] = G.layout("kk")

visual_style["vertex_size"] = 20
visual_style["vertex_label"] = G.vs["name"]
visual_style["label_size"] = .5

visual_style["edge_width"] = [1 for e in G.es]


visual_style["bbox"] = (600, 600)
visual_style["margin"] = 20


# ig.plot(G, **visual_style)



In [ ]:

	ca1_ed	ca2_ed	ca3_ed	ca4_ed	ca5_ed	ca6_ed	ca7_ed	ca8_ed	ca9_ed	ca10_ed	ca11_ed	cafc_ed	cadc_ed
ca1_ing	151702	6656	8507	5928	14008	6875	11991	8486	13901	6697	5072	570	5575
ca2_ing	4229	93422	4849	3134	5570	3374	5918	3715	7092	2879	2226	272	3307
ca3_ing	9683	9711	183320	7779	17799	8669	14532	9541	16940	7783	6017	895	6501
ca4_ing	6677	5920	7749	186566	13910	7350	11521	8450	12537	6601	5771	471	5605
ca5_ing	8541	9195	19136	8885	434933	9263	13788	11229	17853	8531	6696	665	7818
ca6_ing	8227	7556	9942	7980	16663	262366	15882	11250	16922	7979	6346	671	5900
ca7_ing	9839	10101	12246	8392	17765	10952	326501	13214	19171	9299	7310	773	7766
ca8_ing	6857	6463	7917	6813	14228	7754	12830	282702	14574	7916	4992	497	5438
ca9_ing	9664	10656	12182	8566	20039	9710	15850	11809	468980	9775	6702	1179	10503
ca10_ing	7021	5248	7746	5983	13616	7149	11674	9973	14871	237590	5536	600	5139
ca11_ing	4870	3765	5938	4397	36164	4751	7755	5378	9070	4277	202262	532	3030
cafc_ing	1004	935	1660	792	1935	912	1638	916	2898	775	738	76185	1764
cadc_ing	3699	4422	4647	3979	7816	3628	6140	4476	7669	3255	1901	591	96275

	ca1	ca2	ca3	ca4	ca5	ca6	ca7	ca8	ca9	ca10	ca11	cafc	cadc
rank
1	ca1	ca2	ca3	ca4	ca5	ca6	ca7	ca8	ca9	ca10	ca11	cafc	cadc
2	ca5	ca9	ca5	ca5	ca3	ca9	ca9	ca9	ca5	ca9	ca5	ca9	ca5
3	ca9	ca7	ca9	ca9	ca9	ca5	ca5	ca5	ca7	ca5	ca9	ca5	ca9
4	ca7	ca5	ca7	ca7	ca7	ca7	ca8	ca7	ca3	ca7	ca7	cadc	ca7
5	ca3	ca3	ca2	ca8	ca8	ca8	ca3	ca3	ca8	ca8	ca3	ca3	ca3
6	ca8	ca1	ca1	ca3	ca6	ca3	ca6	ca10	ca2	ca3	ca8	ca7	ca8
7	ca6	ca8	ca8	ca6	ca2	ca1	ca2	ca6	cadc	ca6	ca1	ca1	ca2
8	ca10	ca6	ca6	ca1	ca4	ca4	ca1	ca1	ca10	ca1	ca6	ca2	ca4
9	ca2	cadc	ca10	ca10	ca1	ca10	ca10	ca4	ca6	ca4	ca4	ca8	ca1
10	ca4	ca4	ca4	ca2	ca10	ca2	ca4	ca2	ca1	ca11	ca10	ca6	ca6
11	cadc	ca10	cadc	ca11	cadc	ca11	cadc	cadc	ca4	ca2	ca2	ca4	ca10
12	ca11	ca11	ca11	cadc	ca11	cadc	ca11	ca11	ca11	cadc	cadc	ca10	ca11
13	cafc	cafc	cafc	cafc	cafc	cafc	cafc	cafc	cafc	cafc	cafc	ca11	cafc