with our free apps for iPhone, iPad and Android

Get StartedAlready have an account? Log In

NLPA An Overview of Information Retrieval
by Thomas Breuel
#
NLPA An Overview of Information Retrieval

##
advanced topics

###
link analysis

###
matrix decompositions and LSA

###
statistical approaches

###
XML retrieval

###
web search basics

###
web crawling and indexes

##
what is "IR" in practice?

###
text vs knowledge

###
what we want

###
what we get

##
given the vector space model

###
term frequencies turn text documents into vectors

###
once in vector form, we can apply regular pattern recognition and neural network models

###
kinds of models

###
very high dimensional

###
connection

###
commonly used

##
getting at relevance

###
search for documents on the inauguration

###
general idea

###
vector space view

###
choosing weights

###
TF-IDF

###
justification?

##
tolerant retrieval

###
with "grep" or "glob", we can search for arbitrary wildcards like /a.*b/ etc.; these are processed sequentially

###
for large scale IR, we need to be more restrictive since we want faster performance

###
wildcards restricted to terms

###
trailing wildcard within a term

###
leading wildcard within a term

###
wildcard in the middle

###
k-gram indexes

###
general approaches

###
specifics

##
"traditional" IR

###
indexes

###
practical considerations

###
structured texts

##
background

###
use cases

###
properties

##
today

###
inverted indexes

###
efficient computations, out of memory computations

###
TF-IDF

###
basic relationship between vector space models and pattern recognition

0.0 stars - reviews
range from 0 to 5