1. INF 2914 Information Retrieval and Web Search
Lecture 3: Parsing/Tokenization/Storage
These slides are adapted from Stanford’s class CS276 / LING 286
Information Retrieval and Web Mining

2. (Offline) Search Engine Data Flow
Parse & Tokenize
Global Analysis
Index Build
Crawler
web page
- Scan tokenized web pages, anchor text, etc - Generate text index
- Parse - Tokenize - Per page analysis
- Dup detection - Static rank comp - Anchor text
- … 1 2 3 4
in background
tokenized web pages
dup table
rank table
anchor text
inverted text index

3. Inverted index 2 4 8 16 32 64 128 Dictionary Brutus Calpurnia Caesar 1
Posting 2 4 8 16 32 64
128
Dictionary
Brutus
Calpurnia
Caesar 1 2 3 5 8 13 21 34 13 16
Postings lists
Sorted by docID (more later on why).

4. Inverted index construction
Documents to
be indexed.
Friends, Romans, countrymen.
Tokenizer
Token stream.
Friends
Romans
Countrymen
Linguistic modules
Modified tokens.
friend
roman
countryman
Indexer
Inverted index.
friend
roman
countryman 2 4 13 16 1

5. Plan for this lecture The Dictionary Storage XML Introduction Parsing
Tokenization
What terms do we put in the index?
Storage
Log structured file systems
XML Introduction

6. Parsing a document What format is it in? What language is it in?
pdf/word/excel/html?
What language is it in?
What character set is in use?
Each of these is a classification problem.
But these tasks are often done heuristically …

7. Complications: Format/language
Documents being indexed can include docs from many different languages
A single index may have to contain terms of several languages.
Sometimes a document or its components can contain multiple languages/formats
French with a German pdf attachment.
What is a unit document?
A file?
An ? (Perhaps one of many in an mbox.)
An with 5 attachments?
A group of files (PPT or LaTeX in HTML)

8. Tokenization

9. Tokenization Input: “Friends, Romans and Countrymen” Output: Tokens
Each such token is now a candidate for an index entry, after further processing
Described below
But what are valid tokens to emit?

10. Tokenization Issues in tokenization: Finland’s capital 
Finland? Finlands? Finland’s?
Hewlett-Packard  Hewlett and Packard as two tokens?
State-of-the-art: break up hyphenated sequence.
co-education ?
the hold-him-back-and-drag-him-away-maneuver ?
It’s effective to get the user to put in possible hyphens
San Francisco: one token or two? How do you decide it is one token?

11. Numbers
3/12/91 Mar. 12, 1991
55 B.C.
B-52
My PGP key is 324a3df234cb23e
Often, don’t index as text.
But often very useful: think about things like looking up error codes/stacktraces on the web
(One answer is using n-grams, lectures 6 and 7)
Will often index “meta-data” separately
Creation date, format, etc.

12. Tokenization: Language issues
L'ensemble  one token or two?
L ? L’ ? Le ?
Want l’ensemble to match with un ensemble
German noun compounds are not segmented
Lebensversicherungsgesellschaftsangestellter
‘life insurance company employee’

13. Tokenization: language issues
Chinese and Japanese have no spaces between words:
莎拉波娃现在居住在美国东南部的佛罗里达。
Not always guaranteed a unique tokenization
Further complicated in Japanese, with multiple alphabets intermingled
Dates/amounts in multiple formats
フォーチュン500社は情報不足のため時間あた$500K(約6,000万円)
Katakana
Hiragana
Kanji
Romaji
End-user can express query entirely in hiragana!

14. Tokenization: language issues
Arabic (or Hebrew) is basically written right to left, but with certain items like numbers written left to right
Words are separated, but letter forms within a word form complex ligatures
استقلت الجزائر في سنة 1962 بعد 132 عاما من الاحتلال الفرنسي.
← → ← → ← start
‘Algeria achieved its independence in 1962 after 132 years of French occupation.’
With Unicode, the surface presentation is complex, but the stored form is straightforward

15. Normalization
Need to “normalize” terms in indexed text as well as query terms into the same form
We want to match U.S.A. and USA
We most commonly implicitly define equivalence classes of terms
e.g., by deleting periods in a term
Alternative is to do asymmetric expansion:
Enter: window Search: window, windows
Enter: windows Search: windows
Potentially more powerful, but less efficient
Execute queries in parallel or do a second pass over the index

16. Normalization: other languages
Accents: résumé vs. resume.
Most important criterion:
How are your users like to write their queries for these words?
Even in languages that have accents, users often may not type them
German: Tuebingen vs. Tübingen
Should be equivalent

17. Normalization: other languages
Need to “normalize” indexed text as well as query terms into the same form
Character-level alphabet detection and conversion
Tokenization not separable from this.
Sometimes ambiguous:
7月30日 vs. 7/30
Morgen will ich in MIT …
Is this
German “mit”?

18. Case folding Reduce all letters to lower case
exception: upper case (in mid-sentence?)
e.g., General Motors
Fed vs. fed
SAIL vs. sail
Often best to lower case everything, since users will use lowercase regardless of ‘correct’ capitalization…

19. Stop words
With a stop list, you exclude from dictionary entirely the commonest words. Intuition:
They have little semantic content: the, a, and, to, be
They take a lot of space: ~30% of postings for top 30
But the trend is away from doing this:
Good compression techniques means the space for including stopwords in a system is very small
Good query optimization techniques mean you pay little at query time for including stop words.
You need them for:
Phrase queries: “King of Denmark”
Various song titles, etc.: “Let it be”, “To be or not to be”
“Relational” queries: “flights to London”

20. Thesauri and soundex Handle synonyms and homonyms
Hand-constructed equivalence classes
e.g., car = automobile
color = colour
Rewrite to form equivalence classes
Index such equivalences
When the document contains automobile, index it under car as well (usually, also vice-versa)
Or expand query?
When the query contains automobile, look under car as well

21. Soundex
Traditional class of heuristics to expand a query into phonetic equivalents
Language specific – mainly for names
E.g., chebyshev  tchebycheff

22. Lemmatization Reduce inflectional/variant forms to base form E.g.,
am, are, is  be
car, cars, car's, cars'  car
the boy's cars are different colors  the boy car be different color
Lemmatization implies doing “proper” reduction to dictionary headword form

23. Stemming Reduce terms to their “roots” before indexing
“Stemming” suggest crude affix chopping
language dependent
e.g., automate(s), automatic, automation all reduced to automat.
for exampl compress and
compress ar both accept
as equival to compress
for example compressed
and compression are both
accepted as equivalent to
compress.

24. Porter’s algorithm Commonest algorithm for stemming English
Results suggest at least as good as other stemming options
Conventions + 5 phases of reductions
phases applied sequentially
each phase consists of a set of commands
sample convention: Of the rules in a compound command, select the one that applies to the longest suffix.

25. Typical rules in Porter
sses  ss
ies  i
ational  ate
tional  tion

26. Other stemmers
Other stemmers exist, e.g., Lovins stemmer
Single-pass, longest suffix removal (about 250 rules)
Motivated by linguistics as well as IR
Full morphological analysis – at most modest benefits for retrieval
Do stemming and other normalizations help?
Often very mixed results: really help recall for some queries but harm precision on others

27. Language-specificity
Many of the above features embody transformations that are
Language-specific and
Often, application-specific
These are “plug-in” addenda to the indexing process
Both open source and commercial plug-ins available for handling these

28. Index Build Flow - Overview
Crawled
documents
Search
indexes
Per document analysis
Global analysis
Indexing

29. Per document analysis Multi-format parsing
Handles different files types (HTML, PDF, PowerPoint, etc)
Multi-language tokenization, stemming, synonyms, user-defined annotations, etc.
Per document analysis is tipically the bottleneck of the index build process
50 times slower than I/O
Indexing can be done at I/O speed

30. Incorporating per document analysis
Per document analysis is much slower than indexing
Store tokenized documents in a scalable document store
Crawled
documents
Document store
Per document analysis

31. Storage

32. Document store Log-structured file system
Only the most recent version of each document is accessible
No in place updates
Documents are grouped into bundles to optimize I/O
Typically built over the file system
3 basic operation modes
Document insertion (during per-document analysis)
Sequential access for index build
Random access during query processing

33. Store design (1/5) Bundle disk layout Header Doc 2 Doc 1
data
timestamp
# docs
hash(URL)
offset
# fields
length
field ID
field ID
Header
Doc 2
Doc 1
attributes
Bundle disk layout
Fixed bundle size (for instance 8MB)
All fields are 64-bit aligned
All fields are binary
Store does not know how to interpret fields
Compression
Fields are tokens, anchor text, URL, shingle, statistics, etc.

34. Store design (2/5)
Document insertion uses a double buffering algorithm and asynchronous I/O
Try to fit as many documents as it can in a bundle
Schedule write for bundle
Start writing the next bundle

35. Store design (3/5) currentBuffer nextBuffer
Store is sequentially scanned during index build and global analysis
A double buffering algorithm with asynchronous I/O is also used here
Return only the newest version of each document
Store is accessed in reverse order
LFS semantics
currentBuffer
nextBuffer
Bundle# 1053
Bundle# 1052

36. Store design (4/5) Store cleanup algorithm
“Smarter” algorithms can be used if we are not I/O bound
Avoid seeks
New documents
bundle
Bloom filter
D1’
Storei+1
D5’ 1
set 1
bundle D6
D1’ 1
bundle
copy
D5’
Storei D6 D3 1
probe D4
bundle 1 D2 * D1
bundle D3 D4 * D5 D2
* garbage collected

37. Bloom Filters (1/2)
Compact data structures for a probabilistic representation of a set
Appropriate to answer membership queries
False positives!

38. Bloom Filters (2/2)
Query for b: check the bits at positions H1(b), H2(b), ..., H4(b).

39. Store design (5/5)
During runtime the summarizer uses the store to fetch the tokens (random access)
Store provides an API call for retrieving a set of documents (e.g. 20) given its bundle number and offset in the file
Internally the store uses a buffer pool for documents
Asynchronous I/O is used for exploiting parallelism from the storage subsystem
Summarizer releases the documents after it is done

40. Storage Issues Performance Fault tolerance Distribution Redundancy
Field compression
Google File System tries to address these issues

41. XML Introduction

42. Preliminaries: XML <conference> <name> PODS </name>
<speaker>
<name> Josifovski </name>
<paper_cnt> 1 </paper_cnt>
</speaker>
<name> Fagin </name>
<paper_cnt> 3 </paper_cnt>
</conference>
root x0
conference x1
name x2 x6
PODS x3
speaker
speaker x8 x4 x5 x7
name
paper_cnt
explain the scenario: names/ conferences etc.
paper_cnt
name
Josifovski 3 1
Fagin

43. Preliminaries: XPath 1.0 Result: { x7 }
/conference[name = PODS]/speaker[paper_cnt > 1]/name
Query
Document
root
root x0
conference
conference x1
name x2
name
explain more, what is $, what is slash,
$ -> root
expand names, put PODS, etc
take out the arrow, remove predicates
remove slashes – only something else
remove animation
two possible corresp. Only one matches
speaker
= PODS
PODS x6 x3
speaker
speaker x8
paper_cnt
> 1 x4 x5 x7
name
name
paper_cnt
paper_cnt
name
Josifovski 3 1
Fagin
Result: { x7 }

44. XML Indexing article section title figure caption “XML”
//title contains(‘Query Processing’) AND
//figure//caption contains(‘XML’)]
“Query Processing”
In an index-based method, 8 tags and text elements need to be verified to process this query (lessons 6 and 7)

45. Position Encoding A1 B1 B2 C1 D1 B3 C2 R (0,7,0) (1,5,1) (2,2,2)
Scheme #1: Begin/End/Level
Begin: preorder position of tag/text
End: preorder position of last descendent
Level: depth
Containment: X contains Y iff
X.begin < Y.begin <= X.end (assuming well-formed) A1 B1 B2 C1 D1 B3 C2 R
(0,7,0)
(1,5,1)
(2,2,2)
(4,4,3)
(5,5,3)
(6,7,1)
(7,7,2)
(3,5,2)

46. Position Encoding A1 B1 B2 C1 D1 B3 C2 R (1) (1.1) (1.1.1) (1.1.2.1)
Scheme #2: Dewey
Position of element E = {position of parent}.n, where E is the nth child of its parent
Containment: X contains Y iff X is a prefix of Y A1 B1 B2 C1 D1 B3 C2 R
(1)
(1.1)
(1.1.1)
( )
( )
(1.2)
(1.2.1)
(1.1.2)

47. Position Encoding Begin/End/Level Dewey Typically more compact
Fewer implementation issues
Dewey
Encodes positions of all ancestors

48. Path Index Path ID /R 1 /R/A 2 /R/A/B 3 /R/A/B/C 4 /R/A/B/D 5 /R/B 6
/R/B/C 7 A1 B1 B2 C1 D1 B3 C2 R
Describe how we actually use path index to speed up processing later…
Maybe point out the obvious usage, e.g., for path queries
Point out how every position (e.g., B2) is associated with a single path id (e.g., 3)
Mention tradeoffs?
Cons: Space of path table AND postings, time to process index
Pros: enables path twig join and virtual iterators, which can have enormous benefits
Path Pattern -> Set of matching path IDs
/R/B -> {6}
//R//C -> {4, 7}

49. Basic Access Path Inverted posting lists
Posting: <Token, Location>
Token = <Term/Tag>
Location = <DocumentID, Position>
Exercise: Create the posting list representation for the following XML document R A1 B3 B1 B2 C2 C1 D1

50. Inverted index 2 4 8 16 32 64 128 Dictionary Brutus Calpurnia Caesar 1
Posting 2 4 8 16 32 64
128
Dictionary
Brutus
Calpurnia
Caesar 1 2 3 5 8 13 21 34 13 16
Postings lists
Sorted by docID (Why on lessons 6/7).

51. Joins in XML Structural (Containment) Joins Twig Joins A || B B || C BD A || B C D A || B C
Many xml queries are branching expressions, therefore efficiently processing twig joins is an important aspect of any xml query processor

52. Resources for today’s lecture
IIR 2
Porter’s stemmer:
Rosenblum, Mendel and Ousterhout, John K. (February 1992). "The Design and Implementation of a Log-Structured Filesystem." ACM Transactions on Computer Systems. 10(1)
XML Introduction (IIR 10)

53. Trabalho 4 - Proposta Google File System
Map Reduce

54. Trabalho 5 - Proposta XML Parsing, Tokenization, and Indexing
JuruXML - an XML retrieval system at INEX'02
Optimizing cursor movement in holistic twig joins. CIKM 2005: