Test Finding Nemo LIWC

In [1]:

    

import pandas as pd
import matplotlib
%matplotlib inline
import matplotlib.pyplot as plt


    

In [14]:

    

liwc_counts = pd.read_csv('../data/subtitles/subtitlesInTSV/finding_nemo_clean_LIWC_slice_counts.tsv', sep='\t', index_col=False)


    

In [15]:

    

print(liwc_counts.head())


    

    




   anger  death  family  first_person_plural  first_person_singular  home  \
0      0      0       0                    2                      2     1   
1      0      0       1                    1                      3     3   
2      0      0       1                    0                      1     0   
3      0      0       1                    1                      2     0   
4      0      0       2                    2                      3     1   

   humans  insight  negative_affect  percept  positive_affect  sadness  \
0       2        3                0        7               10        0   
1       1        2                0        2                5        1   
2       0        0                0        0                2        0   
3       0        3                1        3                5        1   
4       2        5                2        4                8        0   

   social  slice  
0      14      0  
1      12      1  
2       2      2  
3       5      3  
4      21      4  

In [16]:

    

liwc_categories = sorted(filter(lambda x: x != 'slice', liwc_counts.columns))
print(liwc_categories)


    

    




['anger', 'death', 'family', 'first_person_plural', 'first_person_singular', 'home', 'humans', 'insight', 'negative_affect', 'percept', 'positive_affect', 'sadness', 'social']

In [17]:

    

C = len(liwc_categories)
cols = 5
rows = int(C / cols) + 1
size = 4
plt.figure(figsize=(cols * size, rows * size))
slices = liwc_counts['slice']
for i, c in enumerate(liwc_categories):
    plt.subplot(rows, cols, i+1)
    counts = liwc_counts[c]
    plt.scatter(slices, counts)
    plt.title(c)
    plt.ylim(counts.min(), counts.max()+1)
plt.tight_layout()
plt.show()


    

In [31]:

    

# replicate bubble plot
liwc_categories = list(reversed(liwc_categories))
plt.figure(figsize=(10,10))
ypos = range(len(liwc_categories))
all_ypos = list(reduce(lambda x,y: x+y, [ypos for _ in range(len(slices))]))
print(len(all_ypos))
all_slices = list(reduce(lambda x,y: x+y, [slices.tolist() for _ in ypos]))
print(len(all_slices))
plt.scatter(all_slices, all_ypos, s=liwc_counts[liwc_categories]**2)
plt.yticks(ypos, liwc_categories)
plt.show()


    

    




780
780






    

In [32]:

    

# load actual words
liwc_words = pd.read_csv('../data/subtitles/subtitlesInTSV/finding_nemo_clean_LIWC_slice_words.tsv', sep='\t', index_col=False)


    

In [34]:

    

liwc_words.fillna('', inplace=True)
liwc_words.head()


    

    
Out[34]:






  

    

      

      
anger
      
death
      
family
      
first_person_plural
      
first_person_singular
      
home
      
humans
      
insight
      
negative_affect
      
percept
      
positive_affect
      
sadness
      
social
      
slice
    
  
  

    

      
0
      

      

      

      
our we
      
i my
      
neighborhood
      
kids man
      
believe know think
      

      
beautiful eyes honey look said see view
      
amazing awesome beautiful best better easy goo...
      

      
he honey kids man neighborhood our said talkin...
      
0
    
    

      
1
      

      

      
parents
      
we
      
i me my
      
bedroom house window
      
kids
      
remember think
      

      
look see
      
cutie fine like ok well
      
miss
      
everybody excuse he kids met name parents thei...
      
1
    
    

      
2
      

      

      
daddy
      

      
i
      

      

      

      

      

      
ok promise
      

      
daddy you
      
2
    
    

      
3
      

      

      
dad
      
we
      
i me
      

      

      
answer feel question
      
wrong
      
brush feel see
      
excited fine lucky ok sure
      
missed
      
dad question tell we you
      
3
    
    

      
4
      

      

      
dad mom
      
us we
      
i me my
      
door
      
boy guys
      
know knowledge learn remember wonder
      
doubt shocking
      
look sandy says see
      
funnier funny joke like ok ready safe well
      

      
ask boy bye dad excuse give guys he hey him hi...
      
4
    
  

In [36]:

    

for c in liwc_categories:
    print(c)
    print(liwc_words[c].head())


    

    




social
0    he honey kids man neighborhood our said talkin...
1    everybody excuse he kids met name parents thei...
2                                            daddy you
3                             dad question tell we you
4    ask boy bye dad excuse give guys he hey him hi...
Name: social, dtype: object
sadness
0          
1      miss
2          
3    missed
4          
Name: sadness, dtype: object
positive_affect
0    amazing awesome beautiful best better easy goo...
1                              cutie fine like ok well
2                                           ok promise
3                           excited fine lucky ok sure
4           funnier funny joke like ok ready safe well
Name: positive_affect, dtype: object
percept
0    beautiful eyes honey look said see view
1                                   look see
2                                           
3                             brush feel see
4                        look sandy says see
Name: percept, dtype: object
negative_affect
0                  
1                  
2                  
3             wrong
4    doubt shocking
Name: negative_affect, dtype: object
insight
0                      believe know think
1                          remember think
2                                        
3                    answer feel question
4    know knowledge learn remember wonder
Name: insight, dtype: object
humans
0    kids man
1        kids
2            
3            
4    boy guys
Name: humans, dtype: object
home
0            neighborhood
1    bedroom house window
2                        
3                        
4                    door
Name: home, dtype: object
first_person_singular
0       i my
1    i me my
2          i
3       i me
4    i me my
Name: first_person_singular, dtype: object
first_person_plural
0    our we
1        we
2          
3        we
4     us we
Name: first_person_plural, dtype: object
family
0           
1    parents
2      daddy
3        dad
4    dad mom
Name: family, dtype: object
death
0    
1    
2    
3    
4    
Name: death, dtype: object
anger
0    
1    
2    
3    
4    
Name: anger, dtype: object

In [ ]: