Effective XML Keyword Search with Relevance Oriented Ranking

Effective XML KeywordEffective XML Keyword
Search with RelevanceSearch with Relevance
Oriented RankingOriented Ranking
Zhifeng Bao, Tok Wang Ling, Bo Chen, Jiaheng Lu
1

Introduction
• XML Keyword search
– Inspired by IR style keyword search on the
web
– Enables user to access information in XML
database
– XML data modeled as a rooted, labeled tree
– Recent research efforts
• Efficiency
• Effectiveness
2

Effectiveness
• Capture user’s search intention
– Identify the target that user intends to search for
– Infer the predicate constraint that user intends to
search via
• Result ranking
–Rank the query results according to their
objective relevance to user search intention
3

State of the Art
• Search semantics design
– LCA (Lowest Common Ancestor)
• Node v is a LCA of keyword set K={w1, w2,…,wk} if the sub-tree
rooted at v contains at least one occurrence of all keywords in K,
after excluding the sub-elements that already contain all
keywords in K
– SLCA (Smallest LCA)
• Node v is a SLCA of keyword set K={w1, w2,…,wk} if
– (1) v is a LCA of K
– (2) no proper descendant of v is LCA of K
– XSeek
• Infers the search intention based on the concept of objects and
an analysis of the matching between keyword and data node
4

State of the Art (cont)
• Efficient result retrieval
– Designed based on a certain search semantics
– XKSearch, Multiway SLCA etc.
• Result ranking
– XRANK, XKSEarch, EASE
– They only consider
• Structural compactness of matching results
• Keyword proximity
• Similarity at node level
5

Problems Unaddressed
• Not address the user search intention
adequately!
– Meaningfulness of query result
• SLCA is less meaningful in many cases
– Keyword Ambiguity Problems
1. A keyword can appear both as an xml node type and as
the text value of some other nodes
2. A keyword can appear in the text values of different xml
node types and carry different meaningsNeither SLCA nor Xseek can well address keyword ambiguity
6

Meaningfulness
• Keyword query “rock music”
– Search intention: find customers interested in “rock music”
– C3
– SLCA returns: interest node of C3
customers
storeDB
books
... ...book
title publisherID
authors
author
“B2”
...
“Edward Martin”
“Sophia Jones”
author
customer
ID
name
interest
interests
...
“art”“Rock Davis”
“C4”
...
“Daniel Jones”
“John Williams”
book
title
...ID
authors
author
“B1”
author
“Art of Customer
Interest Care”
customer
ID
name
address
interest
street
city
interests
contact
no.
“1”
“Art Street”
...
...
“fashion”
“Mary Smith”
“C1”
customer
ID
name
interest
interests
“rock music”
“Art Smith”
“C3”
purchase
purchases
customer
ID name
interest
interests
“street art”“John Martin”
“C2”
...
......
... name
“Oxford”
Problems
7

Keyword Ambiguity
• Q = “customer, interest, art”
– Ambiguity 1: customer, interest; Ambiguity 2: art
– Intention: find customer whose interest is art
– less relevant or irrelevant result to be returned also --- C1,C3, B1’s title
customers
storeDB
books
... ...book
title publisherID
authors
author
“B2”
...
“Edward Martin”
“Sophia Jones”
author
customer
ID
name
interest
interests
...
“C4”
...
“Daniel Jones”
“John Williams”
book
title
...ID
authors
author
“B1”
author
“Art of Customer
Interest Care”
customer
ID
name
address
interest
street
city
interests
contact
no.
“1”
“Art Street”
...
...
“fashion”
“Mary Smith”
“C1”
customer
ID
name
interest
interests
“rock music”
“Art Smith”
“C3”
purchase
purchases
customer
ID name
interest
interests
“C2”
...
......
... name
“Oxford”
8
Problems

Keyword Ambiguity (cont)
• Q = “customer, art”
– “art” can be the value of interest node(C2, C4), name node(C3), or
street node of customer(C1), or title node of book(B1)
– “customer” can be tag name of customer node, or (part of) value of
title of(B1)
- How to rank C1 to C4 and B1?
customers
storeDB
books
... ...book
title publisherID
authors
author
“B2”
...
“Edward Martin”
“Sophia Jones”
author
customer
ID
name
interest
interests
...
“C4”
...
“Daniel Jones”
“John Williams”
book
title
...ID
authors
author
“B1”
author
“Art of Customer
Interest Care”
customer
ID
name
address
interest
street
city
interests
contact
no.
“1”
“Art Street”
...
...
“fashion”
“Mary Smith”
“C1”
customer
ID
name
interest
interests
“rock music”
“Art Smith”
“C3”
purchase
purchases
customer
ID name
interest
interests
“C2”
...
......
... name
“Oxford”
9
Problems

Objectives & Challenges
• Challenges
I. How to decide which sub-tree(s) with appropriate node types can
capture user desired information
II. How to return sub-trees of an appropriate size (i.e. contain enough
but non-overwhelming information)
III. How to rank those sub-trees by their relevance
• Address the below as a single problem
– Search intention identification
– Query result retrieval
– Result ranking
– Extend original TF*IDF from text database to XML database,
while capture the hierarchical structure of XML data
10

Challenges
Difficulty in applying TF*IDF to XML
XML DB carries semantic information while text DB
contains pure text information. XML TF*IDF must be
aware of the underlying semantics.
All contents of XML data are stored in leaf nodes only
What is analogy of “flat document” in XML?
o Sub-tree classified according to its prefix path
Normalization factor is not simply the size of sub-tree
o Structure of sub-trees may also infest the ranks
11

TF*IDF Recap
• Rule 1: A keyword appearing in many documents should
not be regarded as more important than a keyword
appearing in a few. --- IDF
• Rule 2: A document with more occurrences of a query
keyword should not be regarded as less important for
that keyword than a document that has less. --- TF
• Rule 3: A normalization factor is needed to balance
between long and short documents
– as Rule 2 discriminates against short documents which may
have less chance to contain more occurrences of keywords.
12

Our Approach
– Extend IR-style keyword search techniques (like TF*IDF)
from text database to XML database, in order to capture the
hierarchical structure of xml document
• by analyzing the knowledge of statistics of underlying XML data
– Major Contributions
1. Identify user’s desired search-for node and search-via node(s) in
a heuristic way
 Define XML TF (term frequency) and XML DF (document frequency)
 Confidence Formulas for search for/via candidates
2. Define XML TF*IDF Similarity
 Propose 3 guidelines specifically for xml keyword search
 Take keyword ambiguity problems into account
3. Design a Keyword Search Engine XReal 13

Data Model
• Node type - Two nodes are of same node type if they share the same
prefix path
/storeDB/customers/customer/name vs.
/storeDB/books/book/publisher/name
customers
storeDB
books
... ...book
title publisherID
authors
author
“B2”
...
“Edward Martin”
“Sophia Jones”
author
customer
ID
name
interest
interests
...
“C4”
...
“Daniel Jones”
“John Williams”
book
title
...ID
authors
author
“B1”
author
“Art of Customer
Interest Care”
customer
ID
name
address
interest
street
city
interests
contact
no.
“1”
“Art Street”
...
...
“fashion”
“Mary Smith”
“C1”
customer
ID
name
interest
interests
“rock music”
“Art Smith”
“C3”
purchase
purchases
customer
ID name
interest
interests
“C2”
...
......
... name
“Oxford”
• Value node – text values contained in leaf node
• Structural node
 Single-valued node type, multi-valued node type
 Grouping type – all its children are of same multi-valued type
14

XML TF and IDF
• XML DF (document frequency)
– The number of T-typed nodes that contain keyword
k in their sub-trees in XML database.
• Granularity of similarity measurement is sub-trees of
certain node type T
• XML TF (term frequency)
– The number of occurrences of a keyword k in a
given value node a in XML database.
T
kf
,a kf
15

Infer the desired search-for node
• Guidelines: A node type T is considered as a desired
search for node if
1. T is intuitively related to every query keyword
2. XML nodes of type T should be informative enough to contain
enough relevant information
3. XML nodes of type T should be not overwhelming to contain too
much irrelevant information
• Confidence of T as the search for node w.r.t. query q.
• product instead of sum is used to follow 1st
guideline
• log part designed to follow 3rd
guideline
• exponential part designed to follow 2nd
guideline
• r is a decay factor in (0,1].
( )
( , ) log (1 )*T depth T
for e k
k q
C T q f r
∈
= +∏
16

Infer the Search-Via Nodes
• Infer structural node to search via
– Structural node n is a good candidate if it is related to as many
(but not necessarily all) keywords as possible
• Search via node type normally is not unique
• Infer individual value node to search via
– Statistics alone is not adequate to infer the likelihood of a value
node as (part of) search via node
– Capture keyword co-occurrence
( , ) log (1 )T
via e k
k q
C T q f
∈
= + ∑
17

customers
storeDB
books
... ...book
title publisherID
authors
author
“B2”
...
“Edward Martin”
“Sophia Jones”
author
customer
ID
name
interest
interests
...
“C4”
...
“Daniel Jones”
“John Williams”
book
title
...ID
authors
author
“B1”
author
“Art of Customer
Interest Care”
customer
ID
name
address
interest
street
city
interests
contact
no.
“1”
“Art Street”
...
...
“fashion”
“Mary Smith”
“C1”
customer
ID
name
interest
interests
“rock music”
“Art Smith”
“C3”
purchase
purchases
customer
ID name
interest
interests
“C2”
...
......
... name
“Oxford”
• E.g. Q = “ customer, name, rock, interest, art ”
 Easy to find name and interest have high confidence to be the
search via nodes
 But hard to know rock is value of name or interest,
art is value of interest or name
How to differ customer C4 from
C3?
Capture keyword co-occurrence
18

Capture keyword co-occurrence
• Proximity factors for a value node v of type kt
containing keyword k
– Given a query q and a certain value node v, if there are two
keywords kt and k in q, s.t. kt matches the type of an
ancestor node of v and k matches a keyword in v
– In-Query distance
• Distance between keyword k and node type kt in query q
• Favors: kt appears before k
– Structural distance
• Depth distance between v and the nearest kt typed
ancestor node of v
– Value-Type distance
• Max of the above two
19
( )
1
( , , ) 1
( , , , )t
via
tk q ancType v
C q v k
Dist q v k k∈ ∩
= + ∑

Principles of XML keyword search
• Principle 1
– When searching for D-typed nodes via a single-valued type V,
ideally only the values and structures nested in V-typed nodes
can affect the relevance, regardless of the size of other typed
nodes nested in D-typed nodes.
• However, TF*IDF similarity in IR normalizes the relevance score of
each document w.r.t. its size
• Principle 2 – address keyword Ambiguity 2
– When searching for nodes of type D via a multi-valued type V’,
the relevance of a D-typed node which contains a query
relevant V’-typed node should not be affected (i.e. normalized)
too much by other query-irrelevant V’-typed nodes.
• Example: query “art” - C4 should not be less relevant than C1
20

Principles of XML keyword search
• Principle 1 and 2
– Especially useful for interpreting pure keyword query -
find search via node correctly
• Principle 3
– The order of keywords in a query is important to
indicate the search intention
• Incorporate the search via confidence Cvia we defined
before
21

XML TF*IDF Similarity
• To calculate the similarity between the search for
node and the query q
– Base case: similarity between value node a and q
• Apply original TF*IDF directly since a contains keywords
only without any structure
– Recursive case: similarity between structural node n
and q
• Based on similarities of its children c and the confidence
level of c as the node type to search via
( , )similarity q a =
,, *
*
Ta
a kq k
k q a
Ta
q a
W W
W W
∈ ∩
∑
IDF TFNormalization
factor
, ( , , )*ln(1 / (1 ))a a
a
T T
q k via T kW C q a k N f= + +
, ,1 ln( )a k a kW f= + 2
,( )a aT T
q q k
k q
W W
∈
= ∑ 2
,a a k
k a
W W
∈
= ∑
22

XML TF*IDF Similarity (cont.)
• Recursive Case
– Intuition 2. An internal node n is relevant to q, if n has a
child c such that the type of c has high confidence to be
a search via node w.r.t. q (i.e. large Cvia(Tc , q)), and c is
highly relevant to q (i.e. large sim(q, c)).
– Intuition 3. An internal node n is more relevant to q if n
has more query-relevant children when all others being
equal.
( )
( , )* ( , )
( , )
via c
c chd n
q
n
sim q c C T q
similarity q n
W
∈
=
∑
Weighted sum of all n’s
children’s similarity and their
confidence to be the search
via node
Overall weight of node n w.r.t
query q which essentially
plays the role of a
normalization factor 23

Flowchart of answering a query
1. Identify user search intention
– Compute the confidence of all possible candidate
node types and choose desired search for node Tfor
2. Relevance-oriented ranking
– Compute XML TF*IDF similarity in a bottom-up
approach from value nodes containing keywords up to
nodes of type Tfor
– Return a ranked list of sub-trees rooted at nodes of
type Tfor
• If more than one search for node type have comparable
confidence, a ranked list for each search for node is returned
24

Experimental Result
• Data set
– DBLP, XMark, WSU, eBay
• Comparison
– Compare XReal with SLCA, Xseek
• Equipment
– Implement in Java
– Run on 3.6GHz pentium IV, 1 GB memory PC with
Windows XP
– Berkeley DB java edition for storing keyword inverted
lists and keyword frequency table
25

Search Effectiveness
• Accuracy in inferring the search for node
– Conducted by user survey
– Tested queries contain at least one of the two
ambiguity problems
– Conclusion
• XReal works well, especially when the search for
node is not given explicitly in the query
26

Search Effectiveness
• Result effectiveness
– Measured by precision, recall, F-measure
– Observations
• XReal achieves higher precision than SLCA and
Xseek for queries that contain ambiguities
• XReal Performs as well as XSeek when queries
have no ambiguity in XML data
• XReal: Top-100 precision higher than overall
precision
• F-measure also shows good overall effectiveness
of both XReal and XSeek 27

Ranking Effectiveness
• Metrics
– Number of Top-1 answers that are relevant
– Reciprocal Rank (R-Rank)
– Mean Average Precision (MAP)
28

Efficiency & Scalability
• Compare three adoptions of indices for
XReal, and SLCA
– Dup
• Store only the dewey id and XML TF
– DupType
• Stores an extra node type (i.e. its prefix path)
– DupTypeNorm
• Stores an extra normalization factor Wa for value
node
,a kf
29

32
customers
storeDB
books
... ...book
title publisherID
authors
author
...
“Edward Martin”
“Sophia Jones”
author
customer
ID
name
interest
interests
...
...
“Daniel Jones”
“John Williams”
book
title
...
ID
authors
authorauthor
“Art of Customer
Interest Care”
customer
ID
name
address
interest
street
city
interests
contact
no.
“1”
“Art Street”
...
...
“fashion”
“Mary Smith”
“C1”
customer
ID
name
interest
interests
“rock music”
“Art Smith”
purchase
purchases
customer
ID name
interest
interests
...
......
... name
“Oxford”
“C2”
“C3”
“C4”
“B1”
“B2”

Effective XML Keyword Search with Relevance Oriented Ranking

More Related Content

Similar to Effective XML Keyword Search with Relevance Oriented Ranking (20)

Recently uploaded (20)

Effective XML Keyword Search with Relevance Oriented Ranking

Editor's Notes