Text Processing
Rong Jin
1
Indexing Process
2
Processing Text


Converting documents to index terms
Why?

Matching the exact string of characters typed by
the user is too restrictive



i.e., it doesn’t work very well in terms of effectiveness
Not all words are of equal value in a search
Sometimes not clear where words begin and end

Not even clear what a word is in some languages (e.g.,
Chinese, Korean)
3
Tokenizing



Forming words from sequence of characters
Surprisingly complex in English, can be
harder in other languages
Early IR systems:



any sequence of alphanumeric characters of
length 3 or more
terminated by a space or other special character
upper-case changed to lower-case
4
Tokenizing

Example:

1.
2.
“Bigcorp's 2007 bi-annual report showed profits
rose 10%.” becomes
bigcorp s 2007 bi annual report showed profits
rose 10
bigcorp 2007
annual report showed profits
rose
5
Tokenizing

Example:




“Bigcorp's 2007 bi-annual report showed profits
rose 10%.” becomes
“bigcorp 2007 annual report showed profits rose”
Too simple for search applications or even
large-scale experiments
Why? Too much information lost

Small decisions in tokenizing can have major
impact on effectiveness of some queries
6
Tokenizing Problems

Small words can be important in some queries,
usually in combinations


xp, ma, pm, ben e king, el paso, master p, gm, j lo, world
war II
Both hyphenated and non-hyphenated forms of
many words are common

Sometimes hyphen is not needed


e-bay, wal-mart, active-x, cd-rom, t-shirts
At other times, hyphens should be considered either
as part of the word or a word separator

winston-salem, mazda rx-7, e-cards, pre-diabetes, spanish7
speaking
Tokenizing Problems


Special characters are an important part of tags,
URLs, code in documents
Capitalized words can have different meaning
from lower case words


Bush, Apple
Apostrophes can be a part of a word, a part of a
possessive, or just a mistake

rosie o'donnell, can't, don't, 80's, 1890's, men's straw
hats, master's degree, england's ten largest cities,
shriner's
8
Tokenizing Problems

Numbers can be important, including
decimals


Periods can occur in numbers, abbreviations,
URLs, ends of sentences, and other situations


nokia 3250, top 10 courses, united 93, quicktime
6.5 pro, 92.3 the beat, 288358
I.B.M., Ph.D., cs.umass.edu, F.E.A.R.
Note: tokenizing steps for queries must be
identical to steps for documents
9
Tokenizing: SMART System
text_stop_file
/amd/beyla/d/smart.11.0/lib/common_words
database
temp_dir
"."
""
trace
0
global_trace_start
-1
global_trace_end
2147483647
global_trace_file
""
global_start
-1
global_end
2147483647
global_accounting
0
# Indexing Locations
doc_loc
query_loc
qrels_text_file
""
query_skel
""
# DEFAULTS FOR INDEXING PROCEDURES AND FLAGS
index_pp
index.index_pp.index_pp
index_vec
index.vec_index.doc
query.index_vec
index.vec_index.query
preparse
index.preparse.generic
next_vecid
index.next_vecid.next_vecid
query.next_vecid
index.next_vecid.next_vecid_1
addtextloc
index.addtextloc.add_textloc
query.addtextloc
index.addtextloc.empty
token_sect
index.token_sect.token_sect
interior_char
"'.@!_"
interior_hyphenation false
endline_hyphenation true
parse.single_case
false
makevec
index.makevec.makevec
expand
index.expand.none
weight
convert.weight.weight
store
index.store.store_vec
doc.store
index.store.store_aux
# DEFAULTS FOR DATABASE FILES AND MODES
dict_file
dict
doc_file
""
textloc_file
textloc
query.textloc_file
""
inv_file
inv
query_file
""
qrels_file
""
collstat_file
""
rwmode
SRDWR
rmode
SRDONLY
rwcmode
SRDWR|SCREATE
## DEFAULTS FOR DOCUMENT DESCRIPTIONS
#### GENERIC PREPARSING
num_pp_sections
0
pp_section.string
""
pp_section.section_name
pp_section.action
copy
pp_section.oneline_flag
false
pp_section.newdoc_flag
false
pp.default_section_name
pp.default_section_action
discard
pp_filter
""
#### PARSING INPUT
index.section.name
none
index.section.ctype -1
method
index.parse_sect.none
token_to_con
index.token_to_con.dict
stem_wanted
1
stemmer
index.stem.triestem
stopword_wanted
1
stopword
index.stop.stop_dict
stop_file
common_words
thesaurus_wanted
0
text_thesaurus_file ""
#### PARSING OUTPUT
index.ctype.name
none
con_to_token
index.con_to_token.contok_dict
store_aux
convert.tup.vec_inv
weight_ctype
convert.wt_ctype.weight_tri
doc_weight
nnn
query_weight
nnn
sect_weight
nnn
num_sections
num_ctypes
0
0
## DEFAULTS FOR COLLECTION CREATION
stop_file_size
1501
dict_file_size
30001
## DEFAULTS FOR CONVERSION
vec_inv.mem_usage
4194000
vec_inv.virt_mem_usage 50000000
in
""
out
""
deleted_doc_file
""
#########################################
## DEFAULTS FOR RETRIEVAL
## Procedures
get_query
retrieve.get_query.get_q_user
user_qdisp
retrieve.user_query.edit_skel
qdisp_query
retrieve.query_index.std_vec
get_seen_docs
retrieve.get_seen_docs.empty
coll_sim
retrieve.coll_sim.inverted
inv_sim
retrieve.ctype_coll.ctype_inv
seq_sim
retrieve.ctype_vec.inner
vecs_vecs
retrieve.vecs_vecs.vecs_vecs
vec_vec
retrieve.vec_vec.vec_vec
retrieve.output
retrieve.output.ret_tr
rank_tr
retrieve.rank_tr.rank_did
sim_ctype_weight
1.0
eval.num_queries
0
num_wanted
10
## Files
tr_file
""
rr_file
""
seen_docs_file
""
run_name
""
#########################################
## DEFAULTS FOR FEEDBACK
## Procedures
feedback
feedback.feedback.feedback
feedback.expand
feedback.expand.exp_const
feedback.occ_info
feedback.occ_info.ide
feedback.weight
feedback.weight.ide
feedback.form
feedback.form.form_query
feedback.num_expand
15
feedback.num_iterations 0
feedback.alpha
1.0
feedback.beta
1.0
feedback.gamma
0.5
#########################################
## DEFAULTS FOR TOP-LEVEL INTERACTIVE DISPLAY AND PRINTING DOCS
indivtext
print.indivtext.text_filter
filter
""
print.format
""
print.rawdocflag
0
print.doctext
print.print.doctext
spec_list
""
verbose
1
max_title_len
68
get_query.editor_only 0
## SUBSET OF TOP-LEVEL ROUTINES (these are ones that are often called
## elsewhere, and substitution of routines may be desired.)
index.doc
index.top.doc_coll
index.query
index.top.query_coll
top.exp_coll
index.top.exp_coll
top.inter
inter.inter
exp_coll
index.top.exp_coll
inter
inter.inter
convert
convert.convert.convert
print
print.top.print
retrieve
retrieve.top.retrieve
## OBSOLETE (only included for backward compatibility)
text_loc_prefix
""
spec_file
./spec
token
index.token.std_token
parse
index.parse.std_parse
# Others
onlyspace_sep 0
interior_slash 0
interior_semicolon 0
An example of configuration file for SMART
## GENERIC PREPARSER
num_pp_sections
pp_section.0.string
pp_section.0.oneline_flag
pp_section.0.newdoc_flag
pp_section.0.action
…
pp_section.8.string
pp_section.8.section_name
pp_section.8.action
## PARSE INPUT
index.num_sections
index.section.0.name
index.section.0.method
index.section.0.word.ctype
index.section.0.proper.ctype
index.section.0.name.ctype
12
"<DOC>"
true
true
discard
"<TEXT>"
w
copy
1
w
index.parse_sect.full
0
0
10
1
Tokenizing Process


First step is to use parser to identify appropriate
parts of document to tokenize
Defer complex decisions to other components




word is any sequence of alphanumeric characters,
terminated by a space or special character, with
everything converted to lower-case
everything indexed
example: Wal-Mart → Wal Mart but search finds
documents with Wal and Mart adjacent
incorporate some rules to reduce dependence on
query transformation components
11
Stopping



Function words (determiners, prepositions)
have little meaning on their own
High occurrence frequencies
Treated as stopwords (i.e. removed)


reduce index space, improve response time,
improve effectiveness
Can be important in combinations

e.g., “to be or not to be”
12
Stopping


Stopword list can be created from highfrequency words or based on a standard list
Lists are customized for applications, domains,
and even parts of documents


e.g., “click” is a good stopword for anchor text
Best policy is to index all words in documents,
make decisions about which words to use at
query time
13
Stopwords in Web Search Engine
14
Stemming

Many morphological variations of words




In most cases, these have the same or very
similar meanings
Stemmers attempt to reduce morphological
variations of words to a common stem


inflectional (plurals, tenses)
derivational (making verbs nouns etc.)
usually involves removing suffixes
Can be done at indexing time or as part of
query processing (like stopwords)
15
Stemming

Generally a small but significant effectiveness
improvement


can be crucial for some languages
e.g., 5-10% improvement for English, up to 50%
in Arabic
Words with the Arabic root ktb
16
Stemming

Two basic types



Dictionary-based: uses lists of related words
Algorithmic: uses program to determine related
words
Algorithmic stemmers

E.g. remove ‘s’ endings assuming plural


e.g., cats → cat, lakes → lake, wiis → wii
Many mistakes: UPS → up
17
Porter Stemmer




Algorithmic stemmer used in IR experiments since
the 70s
Most common algorithm for stemming English
Produces stems not words
Makes a number of errors and difficult to modify
18
Problem with Rule-based Stemming

Sometimes too aggressive in conflation (false
positive)


Sometimes miss good conflations (false negative)


E.g., European/Europe, matrices/matrix,
machine/machinery
Produce stems that are not words


E.g., policy/police, execute/executive, university/universe
E.g., Iteration/iter, general/gener
Corpus analysis can improve or replace a stemmer
19
Krovetz Stemmer

Hybrid algorithmic-dictionary

Word checked in dictionary






If present, either left alone or replaced with “exception”
If not present, word is checked for suffixes that could be
removed
After removal, dictionary is checked again
Produces words not stems
Comparable effectiveness
Lower false positive rate, somewhat higher false
negative
20
Stemming Examples

Original Text
marketing strategies carried out by U.S. companies for their agricultural
chemicals, report predictions for market share of such chemicals, or report
market statistics for agrochemicals.

Porter Stemmer (stopwords removed)
market strateg carr compan agricultur chemic report predict market share
chemic report market statist agrochem

KSTEM (stopwords removed)
marketing strategy carry company agriculture chemical report prediction
market share chemical report market statistic
21
Corpus-Based Stemming
Hypothesis: Word variants that should be conflated co-occur in
documents (text windows) in the corpus

Initial equivalence classes generated by another source


E.g., Porter stemmer
E.g., Common initial prefix or n-grams

New equivalence classes are clusters formed using co-occur
statistics between pairs of words in initial classes

Can be used for other languages
22
Corpus-Based Stemming
(Xu & Croft, 1998)
23
Phrases


Many queries are 2-3 word phrases
Phrases are

More precise than single words


Less ambiguous


e.g., documents containing “black sea” vs. two words
“black” and “sea”
e.g., “big apple” vs. “apple”
In IR, phrases are restricted to noun phrases (or a
sequence of nouns or adjectives followed by
nouns)
24
Phrases


Text processing issue – how are phrases
recognized?
Three possible approaches:



Identify syntactic phrases using a part-of-speech
(POS) tagger
Use word n-grams
Store word positions in indexes and use proximity
operators in queries
25
POS Tagging

POS taggers use statistical models of text to
predict syntactic tags of words

Example tags:


NN (singular noun), NNS (plural noun), VB (verb),
VBD (verb, past tense), VBN (verb, past participle),
IN (preposition), JJ (adjective), CC (conjunction, e.g.,
“and”, “or”), PRP (pronoun), and MD (modal
auxiliary, e.g., “can”, “will”).
Phrases can then be defined as simple noun
groups, for example
26
Pos Tagging Example
27
Pos Tagging Example
28
Example Noun Phrases
29
Word N-Grams


POS tagging too slow for large collections
Simpler definition – phrase is any sequence of n
words – known as n-grams



bigram: 2 word sequence, trigram: 3 word sequence,
unigram: single words
N-grams also used at character level for applications
such as OCR
N-grams typically formed from overlapping
sequences of words


i.e. move n-word “window” one word at a time
“Document will describe…”  “document will” and
“will describe”
30
N-Grams


Frequent n-grams are more likely to be
meaningful phrases
N-grams form a Zipf distribution


Better fit than words alone
Could index all n-grams up to specified length


Much faster than POS tagging
Uses a lot of storage

e.g., document containing 1,000 words would contain
3,990 instances of word n-grams of length 2 ≤ n ≤ 5
31
Google N-Grams



Web search engines index n-grams
Google sample:
Most frequent trigram in English is “all rights
reserved”

In Chinese, “limited liability corporation”
32
Document Structure and Markup


Some parts of documents are more important
than others
Document parser recognizes structure using
markup, such as HTML tags



Headers, anchor text, bolded text all likely to be
important
Metadata can also be important
Links used for link analysis
33
Example Web Page
34
Example Web Page
35
Link Analysis


Links are a key component of the Web
Important for navigation, but also for search




e.g., <a href="http://example.com" >Example
website</a>
“Example website” is the anchor text
“http://example.com” is the destination link
both are used by search engines
36
Anchor Text

Used as a description of the content of the
destination page



i.e., collection of anchor text in all links pointing
to a page used as an additional text field
Anchor text tends to be short, descriptive, and
similar to query text
Retrieval experiments have shown that anchor
text has significant impact on effectiveness
for some types of queries

i.e., more than link analysis
37
Link Quality

Link quality is affected by spam and other
factors


e.g., link farms to increase PageRank
trackback links in blogs can create loops
38
Information Extraction

Automatically extract structure from text


annotate document using tags to identify
extracted structure
Named entity recognition


identify words that refer to something of interest
in a particular application
e.g., people, companies, locations, dates, product
names, prices, etc.
39
Named Entity Recognition


Example showing semantic annotation of text
using XML tags
Information extraction also includes
document structure and more complex
features such as relationships and events
40
Named Entity Recognition

Rule-based

Uses lexicons (lists of words and phrases) that
categorize names


e.g., locations, peoples’ names, organizations, etc.
Rules also used to verify or find new entity names




e.g., “<number> <word> street” for addresses
“<street address>, <city>” or “in <city>” to verify city
names
“<street address>, <city>, <state>” to find new cities
“<title> <name>” to find new names
41
Named Entity Recognition


Rules either developed manually by trial and
error or using machine learning techniques
Statistical



uses a probabilistic model of the words in and
around an entity
probabilities estimated using training data
(manually annotated text)
Hidden Markov Model (HMM) is one approach
42
HMM Sentence Model
• Probabilistic finite state automata
• Each state is associated with a probability distribution over
words (the output)
43
Internationalization



2/3 of the Web is in English
About 50% of Web users do not use English
as their primary language
Many (maybe most) search applications have
to deal with multiple languages


monolingual search: search in one language, but
with many possible languages
cross-language search: search in multiple
languages at the same time
44
Internationalization


Many aspects of search engines are languageneutral
Major differences:




Text encoding (converting to Unicode)
Tokenizing (many languages have no word
separators)
Stemming
Cultural differences may also impact interface
design and features provided
45