Collaborative filtering: prospects and problems

Lecturer : 
Patrik Hoyer
Event type: 
HIIT seminar
Event time: 
2011-01-21 10:15 to 11:00
Place: 
Kumpula Exactum C222
Description: 
Talk announcement:
HIIT Seminar Kumpula, Friday Jan 21 10:15, Exactum C222

SPEAKER:
Patrik Hoyer
Helsinki Institute for Information Technology
University of Helsinki

TITLE:
Collaborative filtering: prospects and problems

ABSTRACT:
Today's abundance of information and of consumer choice in a variety of
domains is making recommendation systems increasingly important in 
helping users find the items they are interested in. Such recommendation 
systems are often based on the idea of collaborative filtering, 
effectively utilizing the correlations among ratings explicitly or 
implicitly provided by users of the system, and are already in 
widespread use in recommending books, movies, songs, etc. I'll consider
the prospects of this technology and in particular discuss some open 
problems that need to be solved for such systems to realize their full
potential. In addition, I will focus on the possibilities and obstacles 
when applying such techniques to the scientific research literature. 


Welcome!
--Matti Järvisalo


HIIT SEMINAR TENTATIVE SCHEDULE Spring 2011
-------------------------------------------
Jan 21: Patrik Hoyer 
Jan 28: *** free ***
Feb  4: Jukka Suomela
Feb 11: André Schumacher
Feb 18: *** free ***
Feb 25: Petteri Kaski
Mar  4: *** free ***
Mar 11: *** free ***
Mar 18: Valentin Polishchuk
Mar 25: *** free ***
Apr  1: *** free ***

EIT ICT Labs - what's in it for HIIT researchers?

Lecturer : 
Marko Turpeinen
Event type: 
HIIT seminar
Event time: 
2011-01-10 13:15 to 14:00
Place: 
Computer Science Building, Room A328
Description: 

 

Our next speaker for HIIT Otaniemi seminar series is Marko Turpeinen from the "Digital Content Communities (DCC)" group of the Helsinki Institute for Information Technology HIIT.

 

All ICS@Aalto researchers are also warmly welcome to attend the seminar!

 

HIIT seminar Otaniemi, Monday January 10, 13:15

Location: Computer Science building, Room A328

 

Marko Turpeinen

Digital Content Communities (DCC) Group

Helsinki Institute for Information Technology HIIT

 

Title:

EIT ICT Labs -  what's in it for HIIT researchers?

 

Abstract:

The mission of EIT ICT Labs is to make Europe a global leader in ICT innovation. It aims at the radical transformation of Europe towards a knowledge-based society and in turning ICT innovation into quality of life. EIT ICT Labs will equip students, researchers, academics and business people with skills for creativity, risk taking spirit and entrepreneurial capacity, by catalysing the existing activities thus leveraging existing regional, national and EU-level funding instruments.

 

The purpose of this event is to present and discuss the EIT ICT Labs activities from researcher's point of view. What is expected from EIT ICT Labs related activities? What is there to be gained in participating? How is this similar and/or different than other EU grants for research?

 

Welcome!

-- 

Mehmet Gönen

Helsinki Institute for Information Technology HIIT

Aalto University, School of Science

Department of Information and Computer Science

 

 

A combinatorial and integrated method to analyse High Throughput Sequencing reads

Lecturer : 
Eric Rivals
Event type: 
HIIT seminar
Event time: 
2011-01-14 10:15 to 11:00
Place: 
Exactum B222
Description: 

Talk announcement:
HIIT Seminar Kumpula, Friday Jan 14, 10:15 a.m., Exactum B222


The first HIIT Seminar Kumpula talk for the Spring term 2011 takes place
Jan 14. Please notice the lecture hall for this presentation (B222).
Forthcoming presentations will take place at C222 as usual.

Welcome!
--Matti Järvisalo

----------------------------------------------------------------------------

SPEAKER:
Eric Rivals
CNRS & Université Montpellier 2
http://www.lirmm.fr/~rivals

TITLE:
A combinatorial and integrated method to analyse High Throughput Sequencing
reads

ABSTRACT:
Next-generation sequencing technologies are presently being used to answer
key biological questions at the scale of the entire genome and with
unprecedented depth. Whether determining genetic or genomic variations,
cataloguing transcripts and assessing their expression levels, finding
recurrent mutations in cancer, identifying DNA-protein interactions or
chromatin modifications, surveying the species diversity in an
environmental sample, all these tasks are now tackled with High Throughput
Sequencing (HTS). For genomics and transcriptomics data sets, the current
paradigm of analysis of large read sets consists in
1. mapping the reads to a reference genome contigously allowing as many
differences as one expects to be necessary to accomodate sequence errors
and small polymorphisms;
2. using uniquely mapped reads to determine covered genomic regions, either
for computing a local coverage to predict SNPs and filter out sequence
errors (cf. program ERANGE), or for delimiting expressed exons
approximately (with RNA-seq; cf. programs TopHat GMORSE),
3. re-aligning unmapped reads, which were not mapped contigously at step
one, to reveal exon boundaries or larger indels.
As shown by the results of approaches following this paradigm, a number of
pitfalls/drawbacks must be accomodated: mapping errors induce false
predictions at further steps, indels larger than 4 bp are not handled, the
impossibility to distinguish SNPs from sequence errors at mapping stage,
the lack of precision on exon boundaries, etc.

On the other hand, we have developped an exact mapper, called MPSCAN, for
short reads (Rivals et al. 09), and analysed its performance in detecting
uniquely mapped regions in function of tag length (Philippe et al. 09). We
could show that one can estimate depending on the genome length, a length k
of substring that will in average point to a single genomic location.
Building on this work, we have conceived a new approach to analyse nowadays
longer reads (> 50 bp). We record for all the k-mers along the read their
matching genomic positions and number of occurrences in the reads, and then
analyse jointly these profiles to determine whether a read can be mapped
contigously or detect multiple causes of alignment disruption: large
indels, introns, rearrangements. In this talk, we will present this
procedure, the underlying data structures, show that it distinguishes SNP
from sequence errors, and allies sensitivity and specificity in the
prediction of exon boundaries, indels, and rearrangements.

Work in collaboration with M. Salson (U. Rouen), N. Philippe et T. Commes
(U. Montpellier 2).

Related publications:
 
  * Using reads to annotate the genome: influence of length, background
distribution, and sequence errors on prediction capacity
  N. Philippe*, A. Boureux*, L. Bréhèlin, J. Tarhio, T. Commes, E. Rivals
  Nucleic Acids Research (NAR) doi:10.1093/nar/gkp492; 2009.
  * MPSCAN: fast localisation of multiple reads in genomes
  E. Rivals, L. Salmela, P. Kiiskinen, P. Kalsi, J. Tarhio
  Proc. 9th Workshop on Algorithms in Bioinformatics
  Lecture Notes in BioInformatics (LNBI), Springer-Verlag, Vol. 5724, p.
246-260, 2009.
 

Hiv-haplotype inference using a constraint-based dirichlet process mixture model

Lecturer : 
Prof. Volker Roth
Event type: 
HIIT seminar
Event time: 
2010-12-16 12:15 to 13:00
Place: 
Lecture hall T3, Computer Science building
Description: 

 Joint ICS Forum and HIIT Seminar Otaniemi talk:


          Hiv-haplotype inference using a constraint-based
                  dirichlet process mixture model

                         Prof. Volker Roth
                   Universität Basel, Switzerland

Abstract:

The recent introduction of the Deep Sequencing technology has led to a
drastic increase in sequenced data, but the limited length of these
sequenced reads and non-negligible sequencing errors have also opened
new statistical challenges in analyzing the data.  This work aims at
analyzing deep sequencing data obtained in a single run from
genetically diverse samples, in particular from intra-host HIV
populations: HIV exists in a single infected patient as quasispecies,
a population of evolutionarily related haplotypes.  We present a
constraint-based Dirichlet process mixture model for identifying (an a
priori unknown number of) these haplotypes in a sample and thereby
quantifying the observed genetic diversity, which defines an important
step in administering personalized medication.

Date:  Thursday, 16 December, 2010
Time:  12:15--13:00
Venue: Lecture hall T3, Computer Science building

Welcome!

Usable Security Case Studies

Lecturer : 
Kristiina Karvonen
Event type: 
HIIT seminar
Event time: 
2010-12-13 13:15 to 14:00
Place: 
Computer Science Building, Room A328
Description: 

 Our next speaker for HIIT Otaniemi seminar series is Kristiina Karvonen from the "Usable Security, Privacy and Trust (USec)" group of the Helsinki Institute for Information Technology HIIT. Before her talk, she will also give a short overview on the research areas of the group.


All ICS@Aalto researchers are also warmly welcome to attend the seminar!

HIIT seminar Otaniemi, Monday December 13, 13:15
Location: Computer Science building, Room A328


Kristiina Karvonen
Usable Security, Privacy and Trust (USec) Group
Helsinki Institute for Information Technology HIIT

Title:
Usable Security Case Studies

Abstract:
Usable security is a field of study that aims to combine research in human computer interaction, security, privacy and trust. Areas of research include, for example, research on innovative security or privacy functionality and design, trust formation, field studies on the use of security or privacy technology, and usability evaluations of new or existing security or privacy features and technology. In my talk I will present an overview of the field and shortly present several case studies to give examples of research in this area.

Welcome!
-- 
Mehmet Gönen
Helsinki Institute for Information Technology HIIT
Aalto University, School of Science and Technology 
Department of Information and Computer Science

Pages