...trovi il tuo blog linkato nelle dispense di due università americane (Central Florida e Central Connecticut). Nice!
Il mio post sul workaround per far funzionare Dev-CPP sotto Vista (dio maledica Microsoft per le sue trovate balzane) è diventato un po' il fix de facto in giro per internet, e ha portato più traffico e commenti delle altre centinaia di post che ho realizzato, anche quelle probabilmente più utili come l'implementazione di Aho-Corasick in PHP (la prima open-source, afaik).
Quest'ultima è un fenomeno interessante, in quanto mi ha provocato solamente un'ondata di spam pornografico nei commenti... in giapponese. Qualcuno deve aver linkato il post su un sito del sol levante (che non riesco a recuperare), e da lì è partito un assalto che, dopo quasi un anno, non accenna ad arrestarsi.
Showing posts with label php. Show all posts
Showing posts with label php. Show all posts
Tuesday, 8 February 2011
Monday, 28 June 2010
Aho-Corasick (multiple pattern match) in PHP
When I started programming Mergifier, my "RSS feeds dispenser", I knew I was going to need some kind of terms filtering. It's useless to take into account the most common words, for they don't bring any information.
My initial solution was to split the text into an array of substrings and use array_diff to remove those in my blacklist array. I don't know the asymptotic cost of array_diff algorythm, but using it with plenty of items (featuring long titles, summaries and descriptions) soon appeared a bit too expensive.
Recently I studied a lot of pattern matching algorythms, including the well known "Aho-Corasick" which checks occourrences of multiple patterns "at once" inside a text of length n in O(M+n), being M the total number of characters composing the patterns. Useful for virus signature detection and bioinformatics, I did not find its implementation anywhere, so I did my own!
Download the code
As you could see inside the code, this algorythm is cool because one can precalculate delta/phi transition tables and cache them, brutally speeding up computations on many different texts made of many terms.
If you find this code useful, or if you have any kind of suggestion to improve it, please let me know. :)
My initial solution was to split the text into an array of substrings and use array_diff to remove those in my blacklist array. I don't know the asymptotic cost of array_diff algorythm, but using it with plenty of items (featuring long titles, summaries and descriptions) soon appeared a bit too expensive.
Recently I studied a lot of pattern matching algorythms, including the well known "Aho-Corasick" which checks occourrences of multiple patterns "at once" inside a text of length n in O(M+n), being M the total number of characters composing the patterns. Useful for virus signature detection and bioinformatics, I did not find its implementation anywhere, so I did my own!
Download the code
As you could see inside the code, this algorythm is cool because one can precalculate delta/phi transition tables and cache them, brutally speeding up computations on many different texts made of many terms.
If you find this code useful, or if you have any kind of suggestion to improve it, please let me know. :)
Subscribe to:
Posts (Atom)

