WEKA version: 3.8.1


My data has many document numbers.
The document has many paragraph text records.
Paragraph text records have word list.
A few important paragraphs in group DOCNO are selected.
Please tell me how to process data segment of group of data records.


DOCNO, PARAGRAPH_NO, PARAGRAPH_TEXT, EXTRACT_WORD_LIST, IMPORTANT_PARAGRAPH
000001, 0001, XXX・・・XXX., XXXXX XXXXX XXXXX, 0
000001, 0002, XXX・・・XXX., XXXXX XXXXX XXXXX XXXXX XXXXX, 1
·
000001, nnnn, XXX・・・XXX., XXXXX XXXXX XXXXX XXXXX, 1
·
000002, 0001, XXX・・・XXX., XXXXX XXXXX XXXXX XXXXX XXXXX, 1
000002, 0002, XXX・・・XXX., XXXXX XXXXX XXXXX , 0
·
000002, nnnn, XXX・・・XXX., XXXXX XXXXX XXXXX XXXXX, 1
·
·
Nnnnnn, 0001, XXX・・・XXX., XXXXX XXXXX XXXXX XXXXX XXXXX, 0
Nnnnnn, 0002, XXX・・・XXX., XXXXX XXXXX XXXXX, 0
·
Nnnnnn, nnnn, XXX・・・XXX., XXXXX XXXXX XXXXX XXXXX, 1