I have these tables. In the first table I have the transaction ID and the products bought in that transaction.
TID Products
1 1,2,5
2 2,4
3 2,3
4 1,2,4
5 1,3
6 2,3
7 1,3
8 1,2,3,5
9 1,2,3
What I want to do to this data is count the number of times an itemset occurs in the Products column. For example the itemset {1,2} is found 4 times in the Products column, there is no problem there, but 1,3 in only found 2 times but as it's seen in the Products column it appears 4 times.
ItemSet Sup. count
1,2 4
1,3 2
1,4 0
1,5 0
2,3 4
2,4 2
2,5 1
3,4 0
3,5 1
4,5 0
This is the php code that I used to find matches in the Sup. count column. if I make it find all coincidences in the itemset when I get to two digit numbers like 12 or more it will say that it found 1 and 2 again that is why I placed the boundary. So I am stuck at this place.
$result_support = array_fill ( 0 , $count_itemsets , 0 );
$count_result_array = count($result_array);
for($i = 0; $i < $count_itemsets; $i++){
for($j = 0; $j < $count_result_array; $j++){
$string = $result_array[$j];
if (preg_match('~\\b' . $result_itemset[$i] . '\\b~i', $string, $m)) {
$result_support[$i] = $result_support[$i] + 1;
}
}
}
$array1 = array(
1=> '1,2,5',
2 => '2,4',
3=> '2,3',
4 => '1,2,4',
5 => '1,3',
6=> '2,3',
7 => '1,3',
8 => '1,2,3,5',
9 => '1,2,3');
$produCt =array();
foreach ($array1 as $index =>$val){
$arra = explode(',', $val);
if(count($arra)>2){
for($i=0;$i<count($arra);$i++){
for($j=$i+1;$j<count($arra);$j++){
$produCt[] = $arra[$i].",".$arra[$j];
}
}
}else{
$produCt[] =$val;
}
}
$prdtCount =array_count_values($produCt);
var_dump($prdtCount);
I assume that your required output is
array (size=8)
'1,2' => int 4
'1,5' => int 2
'2,5' => int 2
'2,4' => int 2
'2,3' => int 4
'1,4' => int 1
'1,3' => int 4
'3,5' => int 1