Blog : Data mining apriori algorithm php implementation

Data mining apriori algorithm php implementation

In data mining, association rule mining is more important to deal with. Apriori algorithm is an algorithm in mining association rules most basic.
I rewrite this period exam with apriori algorithm c php, php lovers doing the mining analysis program design inspired.
Only function to achieve, there is no package

$ Transaction ['item1'] = "i1, i2, i5";
$ Transaction ['item2'] = "i2, i4";
$ Transaction ['item3'] = "i2, i3";
$ Transaction ['item4'] = "i1, i2, i4";
$ Transaction ['item5'] = "i1, i3";
$ Transaction ['item6'] = "i2, i3";
$ Transaction ['item7'] = "i1, i3";
$ Transaction ['item8'] = "i1, i2, i3, i5";
$ Transaction ['item9'] = "i1, i2, i3";
$ The c1 = gen_c1 ($ transaction);
$ The l1 = gen_l1 ($ C1, 1);

/ / Generate c1

for ($ k = 1;; $ k + +)
{
$ {'C'. ($ K +1)} = gen_ck ($ {'l'. $ K}, ($ k +1), $ transaction, 2);
$ {'L'. ($ K +1)} = $ {'c'. ($ K +1)};
if (empty ($ {'c'. ($ k +1)}))
{
exit;
}
else
{
print_r ($ {'c'. ($ k +1)});
the echo "
";

}
/ / $ {'L'. ($ K +1)} = gen_lk ($ {'l'. $ K}, ($ k +1));
}
/ / Print_r ($ c1);
the echo "
";
print_r ($ c2);
the echo "
";
print_r ($ c3);
the echo "
";
print_r ($ c4);
the echo "
";
print_r ($ c5);
the echo "
";


function gen_ck ($ l, $ k, $ t, $ sup = 2)
{
The $ array_row = 0;
for ($ i = 0; $ i {
for ($ j = $ i +1; $ j {
/ / Echo $ l [$ i] ['item']. "=>". $ L [$ j] ['item'];
/ / Echo "
;
$ Temp_array = array ();
$ Match = true;
if ($ k> 2)
{
$ Split_array_i = split (',', $ l [$ i] ['item']) ;/ / ck an item in the data decomposition into an array
/ / Print_r ($ split_array_i);
/ / Echo "i
";

$ Split_array_j = split (',', $ L [$ j] ['item']);
/ / The print_r (the $ split_array_j);
/ / Echo "j
";

for ($ i3 = 0; $ i3 <(count ($ split_array_i) -1); $ i3 + +)
{
if ($ split_array_i [$ i3]! = $ split_array_j [$ i3])
{
$ Match = false;
}
}
}
if ($ match)
{
$ Split_array = split (',', $ l [$ i] ['item']) ;/ / ck an item in the data decomposition into an array
for ($ i1 = 0; $ i1 {
array_push ($ temp_array, $ split_array [$ i1]);
}
$ Split_array = split (',', $ L [$ j] ['item']);
for ($ i1 = 0; $ i1 {
array_push ($ temp_array, $ split_array [$ i1]);
}

/ / Array_push ($ temp_array, $ l [$ i] ['item']);
/ / Array_push ($ temp_array, $ l [$ j] ['item']);
$ Temp_array = array_unique ($ temp_array);

sort ($ temp_array);

if ($ k == 2)
{
/ / The print_r (the $ temp_array);
/ / Echo "
;
}
/ / $ {'C'. $ K} [$ array_row] ['item'] = $ temp_array [0]. ",". $ Temp_array [1];
$ Temp_string ='';
for ($ i4 = 0; $ i4 {
$ Temp_string = $ temp_string. $ Temp_array [$ i4]. ',';
}
/ / Echo $ temp_string;
/ / Echo "
;

$ Temp_string = substr ($ temp_string, 0, strlen ($ temp_string) -1) ;/ / go to the tail of a comma, this module can only be removed in this way at the end of a comma
/ / $ {'C'. $ K} [$ array_row] ['item'] = $ temp_string;
/ / $ {'C'. $ K} [$ array_row] ['support'] = 1;
$ C [$ array_row] ['item'] = $ temp_string;
$ C [the $ array_row] ['support'] = 1;
$ Array_row a + +;
}
else
{
/ / Echo "not match";
}
}
}
/ / Print_r ($ {'c'. $ K});
/ / Support statistics
for ($ i = 0; $ i {
$ Temp_support = 0;

for ($ i1 = 1; $ i1 <= count ($ t); $ i1 + +)
{
$ Temp_array = split (',', $ c [$ i] ['item']);
$ Temp_support1 = true;
for ($ i2 = 0; $ i2 {
/ / Echo $ t ['item'. $ I1]. "=>". $ Temp_array [$ i2];
if (strpos ($ t ['item'. $ i1], $ temp_array [$ i2]) === false)
{
$ Temp_support1 = false;
}

}
if ($ temp_support1)
{
$ Temp_support. + +;
}

}
/ / Echo $ temp_support;
/ / Echo "
;
$ C [$ i] ['support'] = $ temp_support;
}
/ / Remove support for the project is less than the minimum confidence
$ Temp_array = array ();
The $ array_row = 0;
for ($ i = 0; $ i {
if ($ c [$ i] ['support']> = $ sup)
{
$ Temp_array [$ array_row] ['item'] = $ c [$ i] ['item'];
$ Temp_array [$ array_row] ['support'] = $ c [$ i] ['support'];
$ Array_row a + +;
}
}
$ C = $ temp_array;
return $ c;
}
the function gen_lk ($ c)
{

}

the function gen_c1 ($ t)
{
The $ array_row = 0;
for ($ i = 1; $ i <= count ($ t); $ i + +)
{
$ Temp_array = array ();
$ Temp_array = split (',', $ t ['item'. $ I]);
foreach ($ temp_array as $ temp_value)
{
if (in_array_two_dimension (& $ the c $ temp_value)) / / value already exists ck array incremented
{
/ / $ C [$ array_row] ['support'] + +;
}
else / / new value into practice
{
$ C [$ array_row] ['item'] = $ temp_value;
$ C [the $ array_row] ['support'] = 1;
$ Array_row a + +;
}
}
}
return $ c;
}
function gen_l1 ($ c1, $ sup = 2)
{
The $ array_row = 0;
$ Temp_array = array ();
for ($ i = 0; $ i {
if ($ c1 [$ i] ['support']> $ sup)
{
$ Temp_array [$ array_row] ['item'] = $ c1 [$ i] ['item'];
$ Temp_array [$ array_row] ['support'] = $ c1 [$ i] ['support'];
$ Array_row a + +;
}
}
return $ temp_array;
}
the function in_array_two_dimension ($ array, $ value)
{
for ($ i = 0; $ i {
if ($ array [$ i] ['item'] == $ value)
{
$ Array [$ i] ['support'] + +;
return true;
}
}
return false;
}