Tipping Point Demo

Transcript

1. things  you   care  about   things  you  can   influence   the  !pping  point   Take  poli5cal  ac5on  [only]  when  it  ma:ers.  
3. The  DATA   Wide  not  deep:       Of  526  bill  votes:     4  sessions  of  Congress;       174  commi:ees;     1156  keywords;       1414  related  bills;       227  bill  sponsors  …       Manually  sorted  keywords   into  the  five  categories   displayed.  
4. The  DATA   Wide  not  deep:       Of  526  bill  votes:     4  sessions  of  Congress;       174  commi:ees;     1156  keywords;       1414  related  bills;       227  bill  sponsors  …       Manually  sorted  keywords   into  the  five  categories   displayed.   Skewed  dataset:   Most  bills  that  make  it  to  a  vote  pass.    
5. •  10-­‐fold  cross  valida>on  for:   •  Determining  overall  precision  and  recall.   •  Looking  for  model  improvement  as  parameter  sets  are  incorporated.   Trim  final  parameter  set  to  ~200  parameters  that  are  important  over  all  10  valida5on   rounds  –  only  these  will  be  used  to  judge  upcoming  bills.   Data  processing  &  Algorithm:    Random  Forest  
6. Data  processing  &  Algorithm:    Random  Forest   I  ul5mately  want  a  classifier,  but  I  model  as  a  regression  to   get  more  insight.     In  my  case,  precision  is  far  more  cri5cal  than  recall.  
7. Bills  that  have  been  voted  on  since  the  site  went  live:   Passed  at  74%   Predicted  not  to  be  close.   Passed  at  58%   Predicted  to  be  a  close  call!   Passed  at  67%   Predicted  not  to  be  close.  
8. Bills  that  have  been  voted  on  since  the  site  went  live:   Passed  at  74%   Military  Construc5on   Authoriza5on  Act   Passed  at  58%   Intelligence   Authoriza5on  Act   Passed  at  67%   Protect  Medical   Innova5on  Act  
9. Bills  that  have  been  voted  on  since  the  site  went  live:   Passed  at  74%   Military  Construc5on   Authoriza5on  Act   Passed  at  58%   Intelligence   Authoriza5on  Act   Passed  at  67%   Protect  Medical   Innova5on  Act  
10. Bills  that  have  been  voted  on  since  the  site  went  live:   Passed  at  58%   Intelligence   Authoriza5on  Act   Bill  details  go  here!   How  this  bill  is  set  apart  from   other  defense  bills  to  weight  it   toward  close  calls.  
11. Erin  Boyle   PhD  in  Physical  Chemistry,   University  of  Wisconsin  
12. *   *  RF  
13. Future  direc>ons:   •  With  more  data,  could  improve   with  topic-­‐based  modeling.