comparing people in array - php

Question

The problem I'm trying to solve is this.
I have an array of people, some of the people repeat in this array, but their details aren't always exactly the same, there are slight variations in their name. So what I'm trying to do is loop through all the names and compare how close the name is, then if that's a match compare their age and hometown.
Later I hope to create a new array and say "this person also has appears with the following ids" ...
What I have so far is a bubble sort, which if I let it run for long enough I'm sure I can get the job done. I'm just looking to see if someone has a better solution?
<?php
$arr = [
[
'id' => '123',
'name' => 'gary strange',
'home' => 'london',
'age' => 23
],
[
'id' => '124',
'name' => 'john jones',
'home' => 'london',
'age' => 45
],
[
'id' => '125',
'name' => 'bob smith',
'home' => 'paris',
'age' => 63
],
[
'id' => '126',
'name' => 'g strange',
'home' => 'london',
'age' => 23
],
[
'id' => '127',
'name' => 'gary strange',
'home' => 'paris',
'age' => 23
],
[
'id' => '128',
'name' => 'g f. strange',
'home' => 'london',
'age' => 23
]
];
for($i = 0; $i < count($arr); $i++) {
echo "Getting details for " . $arr[$i]['name'] . "\n";
for($j = 0; $j < count($arr); $j++) {
if($j == $i) continue;
else{
echo "Comparing to " . $arr[$j]['name'];
$str1 = $arr[$i]['name'];
$str2 = $arr[$j]['name'];
similar_text($str1,$str2,$percent);
echo " - " . $percent . "%\n";
}
}
echo "******\n";
}

Answer 1

It would probably be more efficient to compare only against people who have a similar age and hometown, and THEN do your comparison checks on name. Comparing against everyone first seems to be like it would be extraordinarily slow.
Something like this for your SQL, as you're iterating through each user:
SELECT name, age, hometown
FROM users
WHERE age BETWEEN value1 AND value2
AND hometown LIKE value3
And THEN do your comparison check. Make sure you use iterators -- this will keep your memory management at bay if you're iterating through tens of thousands of people.
This way, for each user, you're probably only comparing against a handful of possible matches (instead of potentially thousands).