Material Detail

Metoda podpornih vektorjev - 1

This video was recorded at Solomon seminar. Dana je zbirka fizikalnih člankov z znanega strežnika www.arXiv.org. Dobimo besedilo člankov, nekaj metapodatkov in graf citatov med članki. Za nekatere članke vemo tudi, kolikokrat so jih uporabniki prenesli s strežnika v prvih dveh mesecih po objavi članka v arhivu. Naloga je napovedati to število prenosov se za neko vnaprej določeno množico preostalih člankov. To je eden od problemov z letošnjega KDD Cup-a, tekmovanja v data miningu, ki ga organizirajo vsako leto v povezavi s konferenco KDD. Napovedi, ki sva jih za ta problem poslala midva z Juretom, so v konkurenci osemnajstih ekip osvojile prvo mesto. Na seminarju bom podrobneje predstavil problem in opisal, kako sva se ga lotila. Napovedovala sva z linearnim modelom, za učenje sva uporabila regresijski SVM, največ časa pa sva posvetila izbiranju in obteževanju atributov pred učenjem. Vec podrobnosti je na naslovu: http://ai.ijs.si/kddcup03/

Keywords:: videolectures, ocwc, oec

Disciplines:

Science and Technology / Computer Science / Programming & Programming Languages

Go to Material

Bookmark / Add to Course ePortfolio

Create a Learning Exercise

Add Accessibility Information

Rate

Add a Comment

Quality

User Rating
Comments
Learning Exercises
Bookmark Collections
Course ePortfolios
Accessibility Info

Report Broken Link
Report as Inappropriate

More about this material

Material Type:: Presentation
Date Added to MERLOT:: February 10, 2015
Date Modified in MERLOT:: February 10, 2015
Author:: Janez Brank, Artificial Intelligence Laboratory, Jožef Stefan Institute
Submitter:: The Open Education Consortium
Primary Audience:: College General Ed, College Lower Division, College Upper Division
Technical Format:: Video

Mobile Compatibility:: Not specified at this time
Language:: Slovenian
Cost Involved:: No
Source Code Available:: No
Creative Commons:: This work is licensed under a Attribution-NonCommercial-NoDerivs 3.0 United States