Comparing the Donors Choose need statements of teachers in the highest poverty and lowest poverty schools

Here are the number of tweets and the top 10 words in each cluster:

table(kfit$cluster)

## 
##    1    2    3    4    5    6    7 
##  283 1196   79   66  311  661   22

print(clusters_df)

##      Cluster 1 Cluster 2     Cluster 3 Cluster 4 Cluster 5 Cluster 6
## 1          kit   student     projector    pencil       set   science
## 2     electric      need           lcd   markers      ipad   include
## 3      circuit   printer     classroom     erase      mini      cost
## 4        robot  computer    multimedia     board      case  shipping
## 5      handson    camera technological     drill     class  material
## 6      machine    laptop      computer     color classroom targetnew
## 7        solar  makerbot        screen     paper       use      math
## 8       magnet   digital        access   sharpen   protect      book
## 9  investigate     print          need  supplied     apple  activate
## 10        knex    tablet      shipping  notebook   educate   handson
##        Cluster 7
## 1           ipod
## 2          touch
## 3          apple
## 4  technological
## 5           flip
## 6      classroom
## 7      camcorder
## 8         device
## 9           need
## 10       student

Here is a plot of the cosine similarity between topics and clusters:

print(plot)