Learning

Talk Abstract:
Statistical relational learning (SRL) alg

orithms learn statistical models
from relational data such as that sto

red in a relational database. In
prior work we introduced view learning
for SRL in which the view of a
relational database can be automaticall

y modified yielding more accurate
statistical models. In this talk I pr

esent SAYU-VISTA an algorithm which
advances beyond the initial view le

arning approach in three ways. First
it learns views that introduce ne

w relational tables rather than merely
new fields for an existing tabl

e of the database. Second new tables or
new fields are not limited to

being approximations to some target concept;
instead the new approach

performs a type of predicate invention. The
new approach avoids the cla

ssical problem with predicate invention of
learning many useless predic

ates by keeping only new fields or tables
(i.e. new predicates) that

immediately improve the performance of the
statistical model. Third ret

ained fields or tables can then be used in the
definitions of further ne

w fields or tables. We evaluate the new view
learning approach on three
relational classification tasks.

Speaker Bio:
V%EDtor Santos Cos

ta is a Professor at the Universidade do Porto. He was
previously a Prof

essor at Universidade Federal do Rio de Janeiro and
twice a visiting pr

ofessor for a year at the University of Wisconsin-Madison.
His research
interests are on the area of Logic Programming implement-
ation where

he maintains and develops the YAP Prolog system and on
Inductive Logic

Programming and Statistical Relational Learning where
he participated