function selectSubstringHistogram($histogram, $inputText, $substringLength, $count)
{
/*
* Return the lowest scored substring from $inputText using a passed histogram
*/
$substringScores = array();
$split = explode(" ", standardizeText($inputText));
//split standardized string into words
$histogramScores = scoreHistogram($histogram);
//score the histogram once to increase speed
//iterate through all possible substrings of the specified length
for ($i = 0; $i < count($split) - $substringLength + 1; $i++) {
$substring = implode(" ", array_slice($split, $i, $substringLength));
//grab a substring of the correct length
$score = 0;
$words = explode(" ", $substring);
foreach ($words as $word) {
$score += $histogramScores[$word];
}
$substringScores[$i] = $score;
}
asort($substringScores);
//sort the frequency array by value but preserve keys
reset($substringScores);
//reset the key pointer so we can iterate correctly
//grab the lowest scored substring
$substring = implode(" ", array_slice($split, key($substringScores), $substringLength));
//grab the $count'th lowest substring (in case of recalculation
for ($i = 0; $i < $count; $i++) {
if (!next($substringScores)) {
//next() returns false at the end of the array
return "";
//if a unique substring is not found, return ""
}
$substring = implode(" ", array_slice($split, key($substringScores), $substringLength));
}
return $substring;
}
<?php
include "includes/common.php";
$string = "this is a test this is a second test";
//1872
$words = explode(" ", $string);
//split our standardized input by spaces
$histogram = array_count_values($words);
//return an arrray of occurances
print_r($histogram);
echo "<br>";
$histogram = scoreHistogram($histogram);
print_r($histogram);
echo "<br>";
//selectSubstringHistogram($histogram, $inputText, $substringLength, $count)
$substring = selectSubstringHistogram($histogram, $string, 4, 0);
echo "<br>{$substring}<br>";
$sql = array();
foreach ($histogram as $word => $count) {
$sql[] = '("' . $word . '", ' . $count . ')';
}
$blah = array_merge(array_keys($histogram), array_values($histogram));
//echo implode(',', $histogram);
print_r($blah);
