I want to display the quinlan decision tree C4.5 but the program shows me a binary tree !!! What is the problem ?

function node=buildtree(rows,scoref)

   if( size(rows,1)==1  || size(rows(:,end),1)==1 || size(unique(rows(:,end)),1)==1 )
       node=decisionnode(-1,{},uniquecounts(rows));
  return
  end
 if nargin <2
        scoref=@entropy;
 end

current_score=scoref(rows);

 % Set up some variables to track the best criteria
    best_gain=0.0;
    best_criteria={};
    best_sets={};
  %  P=uniquecounts(rows);
  % P=cell2mat(reshape(P(:,2),[1,size(P(:,2),1)]));
  %  P=(P/size(rows,1));
  %  SplitInfo=sum(-P.*log(eps+P)/log(2));

 for col=1:size(rows,2)-1
     SplitInfo=0.0;
      % Generate the list of different values in this column
        if iscellstr(rows(:,col))
            is_cell_str=1;
            column_values=sortrows(unique(rows(:,col)),-1);
             Informations=column_values;

        else
            is_cell_str=0;
            column_values=unique(cell2mat(rows(:,col)));
        end
   for v=1:size(column_values,1)
            if is_cell_str
                set=divideset(rows,col,column_values{v});
            else
                set=divideset(rows,col,column_values(v));
            end

            % Information gain
            p=size(set{1,1},1)/size(rows,1);
            SplitInfo=-p.*log(eps+p)/log(2)+SplitInfo;
          % gain=current_score-p*scoref(set{1,1})-(1-p)*scoref(set{1,2});
          train=set{1,1};
          Informations{v,2}=Information(train,scoref)*p;
   end

          Info=current_score-sum( cell2mat(Informations(:,2)'));

          gain_ratio=Info/SplitInfo;
          %GainRatios=GainRatio(rows,current_score);

          %f=find(GainRatios==max(GainRatios));

          if gain_ratio>best_gain && size(set{1,1},1)>0 && size(set{1,2},1)>0
                best_gain=gain_ratio;
                load data_label;
                if is_cell_str
                   % best_criteria={col,column_values{v},data_label{col}};
                    best_criteria={col,data_label{col}};
                else
                    best_criteria={col,column_values(v)};
                end
                best_sets=set;
           end

end

if best_gain>0

   f1=best_sets{1,1}

        trueBranch=buildtree(best_sets{1,1},scoref);
      f=best_sets{1,2}

                   falseBranch=buildtree(best_sets{1,2},scoref);

                            node=decisionnode(best_criteria{1},best_criteria{2},{},trueBranch,falseBranch)
    else
        node=decisionnode(-1,{},uniquecounts(rows))
end

end