java -jar ruleminer.jar [parameters]

Parameters: 
  --input <FILE>               read data from CSV file <FILE>
  --numericAttributes <LIST>   numbers of columns that will be treated as numeric
  --target <LIST>              list of target columns (i.e. to search for rules maximizing
                               impact of that columns). Target columns must be also numeric
                               (see --numericAttributes)
  --antecedent <LIST>          list of non-numeric columns that will be used to create
                               antecedents (i.e. conditions) of the rules
  --negation <BOOL>            whether to use negative conditions too (i.e. x != 1); negative
                               conditions take much more time to traverse
  --maxlen <INT>               maximum length of antecedents
  --trivialThreshold <INT>     if count_rows(condition U {new_item}) / count_rows(condition) 
                               is above given threshold, the condition will not be enlarged
                               by new_item (1 is feasible)
  --improvementThreshold <INT> the required improvement of impact (in %) in order to enlarge
                               the condition (experimental feature, may not be used to obtain
                               all rules with maximum impact)
  --count <INT>                maximum number of generated rules (e.g. 10)
  --useProjections <BOOL>      if true, the algorithm will perform quicker but will consume
                               more memory
  --minCoverage <DECIMAL>      required minimum coverage of the rules (i.e. discard rules with
                               condition that is satisfied by less than <DECIMAL> % of data 
                               rows (0.1 is feasible)