search query: @keyword bioinformatiikka / total: 14
reference: 1 / 14
« previous | next »
Author:Heiskanen, Tomas
Title:Ranking extension for kernelized Bayesian matrix factorization
Kernelöidyn bayesiläisen matriisihajotelman laajennus järjestyslukuihin
Publication type:Master's thesis
Publication year:2014
Pages:32 s. + liitt. 5      Language:   eng
Department/School:Perustieteiden korkeakoulu
Main subject:Informaatiotekniikka   (IL3010)
Supervisor:Kaski, Samuel
Instructor:Muhammad, Ammad-ud-din ; Georgii, Elisabeth
Electronic version URL: http://urn.fi/URN:NBN:fi:aalto-201501221169
OEVS:
Electronic archive copy is available via Aalto Thesis Database.
Instructions

Reading digital theses in the closed network of the Aalto University Harald Herlin Learning Centre

In the closed network of Learning Centre you can read digital and digitized theses not available in the open network.

The Learning Centre contact details and opening hours: https://learningcentre.aalto.fi/en/harald-herlin-learning-centre/

You can read theses on the Learning Centre customer computers, which are available on all floors.

Logging on to the customer computers

  • Aalto University staff members log on to the customer computer using the Aalto username and password.
  • Other customers log on using a shared username and password.

Opening a thesis

  • On the desktop of the customer computers, you will find an icon titled:

    Aalto Thesis Database

  • Click on the icon to search for and open the thesis you are looking for from Aaltodoc database. You can find the thesis file by clicking the link on the OEV or OEVS field.

Reading the thesis

  • You can either print the thesis or read it on the customer computer screen.
  • You cannot save the thesis file on a flash drive or email it.
  • You cannot copy text or images from the file.
  • You cannot edit the file.

Printing the thesis

  • You can print the thesis for your personal study or research use.
  • Aalto University students and staff members may print black-and-white prints on the PrintingPoint devices when using the computer with personal Aalto username and password. Color printing is possible using the printer u90203-psc3, which is located near the customer service. Color printing is subject to a charge to Aalto University students and staff members.
  • Other customers can use the printer u90203-psc3. All printing is subject to a charge to non-University members.
Location:P1 Ark Aalto  2518   | Archive
Keywords:ranking
kernelized Bayesian matrix factorization
probabilistic ranking
variational approximate inference
Plackett-Luce
drug response prediction
bioinformatiikka
kernelöity bayesiläinen matriisihajotelma
lääkevasteen ennustaminen
syöpägenomiikka
todennäköisyyspohjainen järjestyslukumalli
variaationaalinen approksimatiivinen päättely
Abstract (eng):In cancer drug response prediction, the task is to predict the relative responses of a particular drug compared to other drugs on a set of cell lines.
Essentially, the goal is to model the rank of the drugs and not the actual response values.
New machine learning methods are constantly developed for cancer genomics and many of them still lack rank-based models, for instance kernelized Bayesian matrix factorization (KBMF).
A rank-based model can predict the order of responses of drugs and the results can be compared even though the data might have come from different screening platforms.

In this master's thesis a ranking extension for KBMF is derived.
The ranking is based on Plackett-Luce's ranking model.
The variational equations are derived by using approximation techniques used before in multi-class classification and multinomial logistic regression.
The resulting model is validated against KBMF regression with toy data and cancer drug sensitivity data.
Abstract (fin):Syöpälääkkeiden vasteen ennustuksessa tehtävänä on ennustaa tietyn lääkkeen suhteelliset vasteet verrattuna muiden lääkkeiden vasteisiin solulinjoissa.
Pohjimmiltaan tavoitteena on mallintaa lääkevasteiden järjestys eikä itse vastearvoja.
Uusia koneoppimismenetelmiä kehitetään jatkuvasti syövän genomiikan tutkimusta varten, ja monista menetelmistä puuttuu edelleen järjestyslukuihin perustuvia malleja, esimerkiksi kernelöidylle bayesiläiselle matriisihajotelmalle (KBMF).
Järjestyslukuihin perustuva malli voi ennustaa lääkkeiden vasteiden järjestykset, ja tuloksia voidaan verrata vaikka vasteet olisi peräisin eri seulonta alustoista.

Tässä diplomityössä johdetaan järjestyslukujen laajennus KBMF-menetelmälle.
KBMF laajennus järjestyslukuihin perustuu Plackett-Lucen järjestyslukumalliin.
Variaationaaliset yhtälöt on johdettu käyttämällä approksimaatiotekniikoita, joita on ennen käytetty monen luokan luokituksessa ja multinomiaalisessa logistisessa regressiossa.
Johdettua mallia verrataan KBMF-regressiomenetelmään simuloidulla aineistolla ja syöpälääkkeiden vasteaineistolla.
ED:2015-02-08
INSSI record number: 50532
+ add basket
« previous | next »
INSSI