In [5]:

    

#Load libraries
library(ggplot2)
library(dplyr)


    

In [15]:

    

spon_sentiment_df <- na.omit(read.csv("/home/hao/workspace/6thSemester/DataScience/data/SPON_All.csv", fileEncoding="UTF-16"))
head(spon_sentiment_df)


    

    






X
year
month
day
positiv_abs
neutral_abs
negativ_abs
positiv_rel
neutral_rel
negativ_rel

	
1         
2001      
1         
1         
 4        
  4       
 12       
0.02015050
0.01856187
0.05886288
	
2         
2001      
1         
2         
35        
103       
135       
0.01619442
0.04036150
0.05168048
	
3         
2001      
1         
3         
69        
120       
120       
0.03148833
0.05794016
0.05850294
	
4         
2001      
1         
4         
26        
 49       
 93       
0.01867812
0.03444513
0.06457513
	
5         
2001      
1         
5         
45        
102       
133       
0.02115049
0.04602044
0.05967728
	
6         
2001      
1         
8         
42        
100       
100       
0.02182266
0.05149365
0.05300146

In [7]:

    

spon_sentiment_byYear <- spon_sentiment_df %>% 
  group_by(year) %>%
  summarise(sum_positiv_abs = sum(positiv_abs), sum_neutral_abs = sum(neutral_abs), sum_negativ_abs = sum(negativ_abs), sum_positiv_rel = sum(positiv_rel), sum_neutral_rel = sum(neutral_rel), sum_negativ_rel = sum(negativ_rel))
head(spon_sentiment_byYear)


    

    






year
sum_positiv_abs
sum_neutral_abs
sum_negativ_abs
sum_positiv_rel
sum_neutral_rel
sum_negativ_rel

	
2001    
17061   
40026   
 45843  
7.425997
16.55480
21.39019
	
2002    
19963   
46281   
 58506  
7.344617
16.39999
22.20849
	
2003    
20419   
47891   
 58816  
7.652336
16.81190
22.13802
	
2004    
21964   
50029   
 63053  
7.699428
16.59420
22.28493
	
2005    
28573   
66727   
 86556  
7.424398
16.82071
22.16016
	
2006    
35739   
82787   
105757  
7.410265
17.30626
22.47152

In [8]:

    

calc_polarity <- function(positiv, neutral, negativ) {
    sum <- positiv + neutral + negativ
    polarity <- (positiv - negativ) / sum;
    polarity
}

sum_polarity <- mapply(calc_polarity, positiv = spon_sentiment_byYear$sum_positiv_abs, neutral = spon_sentiment_byYear$sum_neutral_abs, negativ = spon_sentiment_byYear$sum_negativ_abs)


    

In [9]:

    

spon_sentiment_byYear <- cbind(spon_sentiment_byYear, sum_polarity)
head(spon_sentiment_byYear)


    

    






year
sum_positiv_abs
sum_neutral_abs
sum_negativ_abs
sum_positiv_rel
sum_neutral_rel
sum_negativ_rel
sum_polarity

	
2001      
17061     
40026     
 45843    
7.425997  
16.55480  
21.39019  
-0.2796269
	
2002      
19963     
46281     
 58506    
7.344617  
16.39999  
22.20849  
-0.3089619
	
2003      
20419     
47891     
 58816    
7.652336  
16.81190  
22.13802  
-0.3020389
	
2004      
21964     
50029     
 63053    
7.699428  
16.59420  
22.28493  
-0.3042593
	
2005      
28573     
66727     
 86556    
7.424398  
16.82071  
22.16016  
-0.3188402
	
2006      
35739     
82787     
105757    
7.410265  
17.30626  
22.47152  
-0.3121859

In [10]:

    

ggplot(spon_sentiment_byYear, aes(x = year, y = sum_polarity)) + geom_line() +  geom_point() + geom_smooth()


    

    




`geom_smooth()` using method = 'loess'






    










    

In [17]:

    

ggplot(spon_sentiment_byYear, aes(x = year)) + 
geom_line(aes(y=sum_positiv_rel/365, color = "sum_positiv_rel")) + 
geom_line(aes(y=sum_neutral_rel/365, color = "sum_neutral_rel")) +
geom_line(aes(y=sum_negativ_rel/365, color = "sum_negativ_rel"))


    

In [13]:

    




    

In [ ]: