I have noticed that weka does not seem to keep into account encoding.
Is it true or I am missing something ?
For instance, the following two lines from my java script:
BufferedReader arrfReader = new BufferedReader(new InputStreamReader(new
Instances idInstances = new Instances(arrfReader);
Instances does not allow me to specify the encoding, with the result that
I end up issing some of the attributes if they are strings with accented
Anyone knows of a solution ?