Friday, December 04, 2009

Suffix Tree Clustering

I recently stumbled upon this very interesting paper:

Web Document Clustering: A Feasibility Demonstration , by Zamir and Etzioni. PDF here

The paper presents an attractive technique for clustering documents using Suffix Trees, the merits of which can be summarized as:

  • The algorithm is incremental, meaning that documents can be added at any time and clusters can be re-evaluated accordingly
  • Does this in linear complexity O(n)
  • Deals with documents as strings, not as Bags of Words, which allows for accounting for context
  • The algorithm actually produces sequences of words that represent each cluster (with no extra cost)
  • Can produce clusters that are not mutually exclusive (fuzzy clusters)
  • and finally the authors claim it can work efficiently on search engine snippets

Suffix Trees are data structures that are very suitable for indexing text documents (for search purposes), and I think it very devious to exploit the same structure for clustering.

22 comments:

Anonymous said...

[B]NZBsRus.com[/B]
Dismiss Laggin Downloads Using NZB Files You Can Instantly Find High Quality Movies, Games, MP3 Singles, Software and Download Them @ Flying Speeds

[URL=http://www.nzbsrus.com][B]Usenet[/B][/URL]

Anonymous said...

Jibe Our Risqu‚ Prices at www.Pharmashack.com, The Elementary [b][url=http://www.pharmashack.com]Online Chemist's rat on [/url][/b] To [url=http://www.pharmashack.com]Buy Viagra[/url] Online ! You Can also Foretell Hairy Deals When You [url=http://www.pharmashack.com/en/item/cialis.html]Buy Cialis[/url] and When You You [url=http://www.pharmashack.com/en/item/levitra.html]Buy Levitra[/url] Online. We Also Set-up a Stupendous Generic [url=http://www.pharmashack.com/en/item/phentermine.html]Phentermine[/url] On account of Your Nutriment ! We Issue up Border of oeuvre distinguish extreme fell apply to [url=http://www.pharmashack.com/en/item/viagra.html]Viagra[/url] and Also [url=http://www.pharmashack.com/en/item/generic_viagra.html]Generic Viagra[/url] !

Anonymous said...

Genial fill someone in on and this post helped me alot in my college assignement. Gratefulness you on your information.

Anonymous said...

Nice brief and this fill someone in on helped me alot in my college assignement. Thanks you on your information.

Anonymous said...

Predilection casinos? contemplate this advanced [url=http://www.realcazinoz.com]casino[/url] refer to and wing it denigrate online casino games like slots, blackjack, roulette, baccarat and more at www.realcazinoz.com .
you can also put up with over and above our detained [url=http://freecasinogames2010.webs.com]casino[/url] orientate at http://freecasinogames2010.webs.com and acquire sincere genially misled !
another pizazz [url=http://www.ttittancasino.com]casino spiele[/url] tendency is www.ttittancasino.com , in forward movement german gamblers, snitch bear in mind of unrestrained online casino bonus.

Anonymous said...

You could easily be making money online in the underground world of [URL=http://www.www.blackhatmoneymaker.com]blackhat training[/URL], It's not a big surprise if you have no clue about blackhat marketing. Blackhat marketing uses alternative or misunderstood ways to build an income online.

Anonymous said...

i surely enjoy all your posting kind, very unique.
don't quit and keep penning for the reason that it just simply truly worth to follow it.
impatient to find out additional of your content, goodbye :)

Anonymous said...

live video dating http://loveepicentre.com/testimonials.php dating dusseldorf

Anonymous said...

alyson evermore ebook free http://audiobookscollection.co.uk/fr/Microsoft-Solutions-Framework-Essentials-Building-Successful-Technology-Solutions/p155001/ rendesvous with rama ebook [url=http://audiobookscollection.co.uk/de/Time-and-Transcendence-Secular-History-the-Catholic-Reaction-and-the-Rediscovery-of-the-Future-Philosophical-Studies-in-Contemporary-Culture/p212136/]kindle dx ebook reader[/url] convrting books to ebook

Anonymous said...

игровые автоматы через интернет эксплорер [url=http://muhouransutechtwei.narod.ru/file255.html]онлайн покер на реальные деньги[/url] скачать игровые автоматы resident , [url=http://muhouransutechtwei.narod.ru/file795.html]игровые автоматы морской бой torrent[/url] онлайн покер в браузере explorer , [url=http://muhouransutechtwei.narod.ru/file480.html]онлайн покер в индии 6 букв[/url] игровые автоматы скачать игры , [url=http://muhouransutechtwei.narod.ru/file705.html]игровые автоматы admiral[/url] интернет казино список

Anonymous said...

гранд казино онлайн играть [url=http://tinateadartiin.narod.ru/archive28.html]онлайн покер казино фараон 888 poker[/url] имитаторы игровых автоматов бесплатно , [url=http://tinateadartiin.narod.ru/archive616.html]игровые автоматы hot[/url] интернет казино чикаго пермь , [url=http://tinateadartiin.narod.ru/archive504.html]интернет казино остров сокровищ отзывы владельцев[/url] онлайн казино мгновенный вывод денег , [url=http://tinateadartiin.narod.ru/archive798.html]скачать бесплатно клуб азартных игр[/url] игровые автоматы онлайн swamp land , [url=http://tinateadartiin.narod.ru/archive350.html]слот автоматы онлайн бесплатно играть[/url]

Anonymous said...

[url=http://edpengotarmokhaz.narod.ru/casino36.html]играть в покер в интернете[/url] игровые автоматы онлайн крейзи фрут , [url=http://edpengotarmokhaz.narod.ru/casino420.html]игровые автоматы онлайн swamp land[/url] игровые автоматы вулкан играть бесплатно и без регистрации , [url=http://edpengotarmokhaz.narod.ru/casino660.html]игровые автоматы слоты онлайн[/url] игровые автоматы columbus цена , [url=http://edpengotarmokhaz.narod.ru/casino60.html]игровые автоматы яндекс бесплатно[/url] онлайн покер калькулятор для сети i poker club , [url=http://edpengotarmokhaz.narod.ru/casino612.html]игровые автоматы клубника играть бесплатно и без регистрации[/url] super jump игровые автоматы

Anonymous said...

игровые автоматы фрукты zte игровой автомат олимпийские игры игровые автоматы советских времен , азартные игры гараж слот автоматы играть бесплатно онлайн ещё игры , игровые автоматы без регистрации однорукий бандит слот автоматы играть бесплатно лягушки , игровые автоматы без регистрации скачать игровой автомат золото партии играть , казино онлайн с контролем честности

Anonymous said...

азартные игры мультиплеер [url=http://framculakrugodjue.narod.ru/doc646.html]интернет рулетка на рубли[/url] азартные игры играть бесплатно без регистрации онлайн , [url=http://framculakrugodjue.narod.ru/doc357.html]лучшие онлайн игроки в покер[/url] игровые автоматы лягушки играть бесплатно онлайн , [url=http://framculakrugodjue.narod.ru/doc612.html]онлайн казино crazy monkey на русском[/url] виртуальное казино 4 дракона , [url=http://framculakrugodjue.narod.ru/doc544.html]интернет рулетка без регистрации на русском[/url] интернет казино онлайн crystal palace

Anonymous said...

german-english translation software http://buyoemsoftware.co.uk/product-36778/SkyCode-WebTrance-3-0-Bilingual dvd data burning software free [url=http://buyoemsoftware.co.uk/it/category-14/Altri?page=3]bma users group software bcr[/url] best no dvd duplicating software
[url=http://buyoemsoftware.co.uk/product-33469/DigitalAnarchy-ToonIt-2-51-for-Photoshop-CS3-x64]DigitalAnarchy ToonIt 2.51 for Photoshop CS3 x64 - Software Store[/url] wedding invitation design software
[url=http://buyoemsoftware.co.uk/fr/product-36682/DBF-Reanimator-1-0][img]http://buyoem.co.uk/image/3.gif[/img][/url]

Anonymous said...

http://www.msn.com/ - [url=http://www.msn.com/]msn[/url] msn
msn

Anonymous said...

Inherhark xaikalitag Skimizmal [url=http://uillumaror.com]iziananatt[/url] psymmentonamn http://gusannghor.com wreddegodet

Anonymous said...

I have been inѵited, and thе business real issue is, is there
а rank order therе? If yоur credіt iѕ аn advantаge, as thеy have
beеn buуing and selling is аll abοut delivering effісiencу to the
workplaсe etiquette discussіon. Internеt marketing isn't just a parts supplier, they are quite simply shocking. Once you have gotten this far, may I suggest you go on a diet, work out how much it will cost you per month.

Here is my website :: internet marketing companies

Anonymous said...

Good web site you have here.. It's hard to find quality writing like yours these days. I really appreciate individuals like you! Take care!!

My homepage www.lepremiersoir.fr

Anonymous said...

Hi! I've been following your web site for a while now and finally got the courage to go ahead and give you a shout out from Houston Tx! Just wanted to tell you keep up the excellent job!

Here is my blog post Http://Dentalrootcanalss.wallinside.Com/

Anonymous said...

After looking at a few of the articles on your website,
I truly like your way of blogging. I bookmarked it to my bookmark
site list and will be checking back soon. Please visit my website too and let me
know your opinion.

Also visit my blog post; dental implant costs

Anonymous said...

free oriental dating sites http://loveepicentre.com/map/ dating chatroom
speed dating in thousand oaks [url=http://loveepicentre.com/contact/]radiometric dating and relative dating[/url] dating services little rock
dating indians in burlington nc [url=http://loveepicentre.com/advice/]divorced and dating[/url] good australian dating site [url=http://loveepicentre.com/user/dimadl/]dimadl[/url] online internet dating utah