Options
Significant Attributes per Label/Class [Solved]
CaptainChaos
Member Posts: 17 
Contributor II
Hi Guys,
so far I just have some experience with "Text Mining". Due to my job i now have to analyse data from another area. I have a spreadsheet with products produced and errors occuring during production. I transfered all the data to spreadsheat in excel which looks like this:
Productid Label Error1 Error2 Errorr3....
1 A 0 2 0
2 B 1 0 1
3 B 0 1 0
. . . . .
. . . . .
. . . . .
My first question is if its ok to put a zero in a cell when the error does not occur or if it would be more appropiate to leave it with out the zero. Second question would be, if some body could tell me how i could find out which the most significant errors/attribute are for one class, so what operator to use.
Thanks in advance for your help ;D
so far I just have some experience with "Text Mining". Due to my job i now have to analyse data from another area. I have a spreadsheet with products produced and errors occuring during production. I transfered all the data to spreadsheat in excel which looks like this:
Productid Label Error1 Error2 Errorr3....
1 A 0 2 0
2 B 1 0 1
3 B 0 1 0
. . . . .
. . . . .
. . . . .
My first question is if its ok to put a zero in a cell when the error does not occur or if it would be more appropiate to leave it with out the zero. Second question would be, if some body could tell me how i could find out which the most significant errors/attribute are for one class, so what operator to use.
Thanks in advance for your help ;D
0
