Presentation is loading. Please wait.

Presentation is loading. Please wait.

S EQUENCE I NDEXING S CHEMES Roman Čížek Erasmus 2687, Nelly Vouzoukidou MET601.

Similar presentations


Presentation on theme: "S EQUENCE I NDEXING S CHEMES Roman Čížek Erasmus 2687, Nelly Vouzoukidou MET601."— Presentation transcript:

1 S EQUENCE I NDEXING S CHEMES Roman Čížek Erasmus 2687, Nelly Vouzoukidou MET601

2 I NTRODUCTION Graph indexes precise Path, (twig only few methods) Sequence indexing schemes Top-down or bottom-up XML document and XML queries in structure-encoded sequences Path and twig

3 T OP -D OWN S EQUENCE I NDEXES : V I ST

4 V I ST – V IRTUAL S UFFIX T REE Top-down Sequence Indexes Represent XML documents and XML queries in structure-encoded sequences Querying XML data is equivalent to finding subsequence matching Avoid to expensive join operations Provides unified index on both content and structure Support dynamic index update B + Trees which are supported in DBMSs

5 DTD OF PURCHASE RECORDS

6 A S INGLE P URCHASE R ECORD

7 P REORDER S EQUENCE OF XML Use capital letters to represent names of elements/attributes Use hash function h(), to encode attribute values into integers v 1 = h(“dell”) v 2 =h(“ibm”) Preorder sequence of XML purchase record example PSNv 1 IMv 2 Nv 3 IMv 4 INv 5 Lv 6 BLv 7 Nv 8 Isomorphic trees may produce different preorder seq. DTD schema embodies linear order of all elements/attributes Without DTD – use lexicographical order

8 S TRUCTURE -E NCODED S EQUENCE Definition: A Structure-Encoded Sequence, derived from a prefix traversal of semi-structured XML document, is a sequence of (symbol, prefix) pairs: D = (a 1,p 1 ), (a 2,p 2 ),…, (a n,p n ) Where a i represents a node in the XML document tree, (of which a 1, …,a n is the preorder sequence), and p i is the path from the root node to node a i.

9 S TRUCTURE -E NCODED S EQUENCE D= (P,ϵ),(S,P),(N,PS),(v 1,PSN),(I,PS),(M,PSI),(v 2,PSIM),(N,PSI), (v 3,PSIN),(I,PSI),(M,PSII),(v 4,PSIIM),(I,PS),(N,PSI),(v 5,PSIN), (L,PS),(v 6,PSL),(B,P),(L,PB),(v 7,PBL),(N,PB),(v 8,PBN)

10 XML Q UERIES IN G RAPH F ORM

11 XML Q UERIES IN P ATH E XPRESSION AND S EQUENCE F ORM Query: Path Expression Structure-Encoded Sequence Q1 : /Purchase/Seller/Item/Manufacturer (P, ϵ)(S,P)(I,PS)(M,PSI) Q2 : /Purchase[Seller[Loc = v 5 ]]/Buyer[Loc = v 7 ] (P, ϵ)(S,P)(L,PS)(v 5,PSL)(B,P)(L,PB)(v 7,PBL) Q3 : /Purchase/*[Loc = v 5 ] (P, ϵ)(L, P)(v 5,P*L) Q4 : /Purchase//Item[Manufacturer = v 3 ] (P, ϵ)(I,P//)(M, P//I)(v 3,P//IM)

12 Q UERYING XML THROUGH S TRUCTURE -E NCODED S EQUENCE M ATCHING Querying XML is equivalent to finding (non-contiguous) subsequence matches Most structural XML queries can be performed through direct subsequence matching Exception: branch has multiple identical child nodes Q 5 =/A[B/C]/B/D Two different sequences (A, ϵ)(B,A)(C,AB)(B,A)(D,AB) (A, ϵ)(B,A)(D,AB)(B,A)(C,AB) Find matches separately and union their result We may find false matches if the indexed documents contain branches with identical child nodes, then we ask multiple queries and compute set difference on result If the query contains a large number of same child nodes under the branch, we can choose disassemble the tree into multiple trees and use join operations to combine their results

13 A LGORITHMS Naïve algorithm RIST – Relationships Indexed Suffix Tree ViST – Virtual Suffix Tree

14 N AÏVE ALGORITHM : S UFFIX -T REE -L IKE STRUCTURE Doc1 : (P, ϵ)( S, P)(N, PS)(v 1, PSN)(L, PS)(v 2, PSL) Doc2 : (P, ϵ)(B, P)(L, PB)(v 2, PBL) Q1 : (P, ϵ)(B, P)(L,PB)(v 2, PBL) Q2 : (P, ϵ)(L, P*)(v 2,P*L)

15 D-A NCESTORSHIP AND S-A NCESTORSHIP D-Ancestorship Ancestor-descendant relationships in original XML tree Element (S,P) is a D-Ancestorship of (L,PS) S-Ancestorship Ancestor-descendant relationships in suffix tree Element (v 1, PSN) is an S-Ancestorship of (L, PS)

16 N AÏVE SEARCH :A NAÏVE ALGORITHM BASED ON SUFFIX TREES

17 RIST – I NDEXING C ONSTRUCTION S-Ancestorship requires additional information Label each suffix tree node x by pair n x prefix traversal order of x in suffix tree size x is total number of descendants of x in suffix tree x …, y … x is S-Ancestor of node y if n y ϵ (n x, n x + size x ] Construct the B + Trees: Tree nodes into the D-Ancestorship B + Tree using (Symbol, Prefix) as keys For all nodes x inserted with the same (Symbol, Prefix) we index them by S-Ancestorship B + Tree, using the n x values of their labels as keys.

18 T HE RIST INDEX STRUCTURE

19 S EARCH : NON - CONTIGUOUS SUBSEQUENCE MATCHING USING B + T REE

20 V I ST – V IRTUAL S UFFIX T REE Dynamic Virtual suffix tree labeling Semantic and statistical clues Dynamic scope allocation without clues

21 D YNAMIC SCOPE ALLOCATION Number of child nodes of x is λ. We allocate 1/ λ of the remaining scope to x’s first child Dynamic scope allocation with λ=2

22 D YNAMIC S COPE OF A S UFFIX T REE N ODE

23 SUB S COPE ( PARENT, E ): CREATE A SUB SCOPE WITHIN THE PARENT SCOPE FOR E

24 I NSERTION INDEX Doc 1 = (P, ϵ)(S,P)(N,PS)(v 1,PSN)(L,PS)(v 2,PSL) Doc 2 = (P, ϵ)(S,P)(L,PS)(v 2,PSL)

25 I NDEX AN XML DOCUMENT

26 EXPERIMENTS - S AMPLE QUERIES Path Expression Dataset Q1 /inproceedings/title DBLP Q2 /book/author[text=‘David’] DBLP Q3 /*/author[text= ‘David’] DBLP Q4 //author[text= ‘David’] DBLP Q5 /book[key=‘books/bc/MaierW88’]/author DBLP Q6 /site//item[location=‘US’]/mail/date[text=‘12/15/1999’] XMARK Q7 /site//person/*/city[text=‘Pocatello’] XMARK Q8 //closed_auction[*[person=‘person1’]]/date[text=‘12/15/1999’] XMARK

27 C OMPARING INDEXING METHODS time in seconds

28 I NDEX STRUCTURE DBLP (301 MB of data) XMARK (52MB of data)

29 C ONCLUSION structure-encoded sequences Sequence matching Avoid expensive join operations Top-down scope allocation method Index structure – B + Tree

30 PRIX: P RUFER S EQUENCES FOR I NDEXING XML

31 Rao & Moon (2006) proposed a new method for indexing XML documents using sequences It uses the same idea as in ViST index: The XML tree is transformed into a sequence and saved in the database Each query is also transformed into a sequence The answer of the query is acquired by performing subsequence matching

32 PRIX: PR UFER SEQUENCES FOR I NDEXING X ML Motivation PRIX architecture Indexing XML documents Querying

33 PRIX: PR UFER SEQUENCES FOR I NDEXING X ML Motivation PRIX architecture Indexing XML documents Querying

34 M OTIVATION : T WIG Q UERIES AND W ILDCARDS Like in ViST, PRIX also tries to efficiently answer twig queries as well as queries containing wildcards (‘*’ any and ‘//’ self or descendant queries) P Q TS Twig query XPath: P/Q[T]/S Query with wildcards XPath: P//Q/S P Q S

35 M OTIVATION : P ROBLEMS IN V I ST I NDEX Memory requirements: In the worst case, ViST requires O(N 2 ) space to index the document A B C D D = (A, ε), (B, A), (C, AB), (D, ABC), (E, ABCD) E Elements in height k appear k times

36 M OTIVATION : P ROBLEMS IN V I ST I NDEX Memory requirements: In the worst case, ViST requires O(N 2 ) space to index the document False positives In many cases, query processing in Vist results in false alarms P Q T R TUS Doc1 = (P, e) (Q, P) (T, PQ) (S, PQ) (R, P) (U, PR) (T, PR) P Q T Q S Doc2 = (P, e) (Q, P) (T, PQ) (Q, P) (S, PQ) P Q TS XPath: P/Q[T]/S Q = (P, e) (Q, P) (T, PQ) (S, PQ)

37 M OTIVATION : P ROBLEMS IN V I ST I NDEX Memory requirements: In the worst case, ViST requires O(N 2 ) space to index the document False positives In many cases, query processing in Vist results in false alarms False negatives Correctly answering a twig query depends on the order the branches are created P F T N G Doc = (P, e) (F, P) (T, PF) (N, P) (G, PN) P NF Xpath: P[N]/F Q = (P, e) (N, P) (F, P) ???

38 M OTIVATION : P ROBLEMS IN V I ST I NDEX Memory requirements: In the worst case, ViST requires O(N 2 ) space to index the document False positives In many cases, query processing in Vist results in false alarms False negatives Correctly answering a twig query depends on the order the branches are created

39 PRIX: PR UFER SEQUENCES FOR I NDEXING X ML Motivation PRIX architecture Indexing XML documents Querying

40 PRIX A RCHITECTURE

41 I NDEXING AND Q UERYING IN PRIX Indexing: The first step is to take as input an XML document and convert it into a sequence This is achieved using Prufer Sequences The sequence is saved in the database in a way equivalent to the one used in ViST It is a Virtual Trie implemented as B+ Trees XML document

42 I NDEXING AND Q UERYING IN PRIX Querying Queries are also transformed to trees and then to Prufer Sequences The query sequence looked up in the document sequence and all matching subsequences are retrieved After this initial filtering, three refinement phases follow XPath Query

43 PRIX: PR UFER SEQUENCES FOR I NDEXING X ML Motivation PRIX architecture Indexing XML documents Querying

44 I NDEXING XML D OCUMENTS The first step is to transform the XML document to the equivalent XML tree Notice that both elements and text values are represented as nodes (the same stands for attributes) The tree is not saved in the database D A BB CC D F E

45 I NDEXING XML D OCUMENTS Then the Prufer Sequence is created from the XML tree A Prufer Sequence is a method proposed by Prufer (1918) that constructs a one-to-one correspondence between a labeled tree and a sequence 8,A 1,B7,B 6,C3,C 2,D 5,E 4,F 8, 3, 7, 6, 6, 7, 8

46 I NDEXING XML D OCUMENTS Prufer Sequences can only be created from trees with numerical labeling, with each node having a unique number Since the XML tree contains string labels (the names of elements etc.) we add an additional label to each node We will use the post-order traversal to name the nodes The prufer sequence can be extracted for any labeling of the tree, but using post-order numbering has some properties that makes the querying process easier

47 I NDEXING XML D OCUMENTS Initial labeling A BB CC D F E 8,A 1,B7,B 6,C3,C 2,D 5,E 4,F

48 I NDEXING XML D OCUMENTS Finding the Prufer Sequence The algorithm to find the Prufer sequence is the following: Find the leaf with the smallest value and delete it. Add the label of its parent to the sequence Repeat until only one node is left In PRIX index, two sequences are held: The actual Prufer Sequence holding the numbers of the labels called Numbered Prufer Sequence: NPS The corresponding sequence holding the actual labels of the nodes of the XML Tree called Labeled Prufer Sequence: LPS

49 I NDEXING XML D OCUMENTS Finding the Prufer Sequence The algorithm to find the Prufer sequence is the following: Find the leaf with the smallest value and delete it. Add the label of its parent to the sequence Repeat until only one node is left 8,A 1,B7,B 6,C3,C 2,D 5,E 4,F NPS : 8, LPS : A,

50 I NDEXING XML D OCUMENTS Finding the Prufer Sequence The algorithm to find the Prufer sequence is the following: Find the leaf with the smallest value and delete it. Add the label of its parent to the sequence Repeat until only one node is left 8,A 7,B 6,C3,C 2,D 5,E 4,F NPS : 8, 3 LPS : A, C 1,B

51 I NDEXING XML D OCUMENTS Finding the Prufer Sequence The algorithm to find the Prufer sequence is the following: Find the leaf with the smallest value and delete it. Add the label of its parent to the sequence Repeat until only one node is left 8,A 7,B 6,C3,C 2,D 5,E 4,F NPS : 8, 3, 7, 6, 6, 7, 8 LPS : A, C, B, C, C, B, A 1,B

52 I NDEXING XML D OCUMENTS Properties Both NPS and LPS have length N-1 (where N is the total number of nodes Due to the fact that we delete one node at a time until only one node is left NPS : 8, 3, 7, 6, 6, 7, 8 LPS : A, C, B, C, C, B, A 8,A 1,B7,B 6,C3,C 2,D 5,E 4,F

53 I NDEXING XML D OCUMENTS Properties Both NPS and LPS have length N-1 (where N is the total number of nodes The i-th element deleted is always the node with label i This helps us find the edges of the tree! (that is the mapping from the NPS to the tree) NPS : 8, 3, 7, 6, 6, 7, 8 LPS : A, C, B, C, C, B, A 8,A 1,B7,B 6,C3,C 2,D 5,E 4,F Deleted node: 1, 2, 3, 4, 5, 6, 7

54 I NDEXING XML D OCUMENTS Properties Both NPS and LPS have length N-1 (where N is the total number of nodes The i-th element deleted is always the node with label i LPS does not contain any leaves NPS : 8, 3, 7, 6, 6, 7, 8 LPS : A, C, B, C, C, B, A 8,A 1,B7,B 6,C3,C 2,D 5,E 4,F Deleted node: 1, 2, 3, 4, 5, 6, 7

55 I NDEXING XML D OCUMENTS Indexes held in the database are The LPS (label prufer sequence) The NPS (numbered prufer sequence) The mapping between the number and the xml label of the leaves of the tree NPS : 8, 3, 7, 6, 6, 7, 8 LPS : A, C, B, C, C, B, A Leaves mapping: 1  B, 2  D, 4  F, 5  E 8,A 1,B7,B 6,C3,C 2,D 5,E 4,F

56 PRIX: PR UFER SEQUENCES FOR I NDEXING X ML Motivation PRIX architecture Indexing XML documents Querying

57 Q UERYING When a query arrives it is also transformed to a prufer sequence Then, an initial filtering is performed The results of the initial filtering are sorted out in order to acquire the correct answer to the query after three more refinement phases. XPath Query

58 Q UERYING : T RANSFORMING A QUERY TO A PRUFER SEQUENCE The same process as in documents is followed For instance if we have the XPath query A[B/C]/D/E/F The query tree is: The NPS and LPS are: NPS(Q) = 2, 6, 4, 5, 6 LPS(Q) = B, A, E, D, A

59 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING Suppose we have the following XML tree (T) of the document: NPS(T) = 15, 3, 7, 6, 6, 7, 15, 9, 15, 13, 13, 13, 14, 15 LPS(T) = A, C, B, C, C, B, A, C, A, E, E, E, D, A

60 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING To find the correct results for the given query we find the subsequences of LPS(Q) inside LPS(T) “A subsequence is any string that can be obtained by deleting zero or more symbols from a given string”

61 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING To find the correct results for the given query we find the subsequences of LPS(Q) inside LPS(T) Each subsequence represents a possible solution in the tree LPS(T) = A, C, B, C, C, B, A, C, A, E, E, E, D, A LPS(Q) = B, A, E, D, A T Q

62 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING To find the correct results for the given query we find the subsequences of LPS(Q) inside LPS(T) Each subsequence represents a possible solution in the tree LPS(T) = A, C, B, C, C, B, A, C, A, E, E, E, D, A LPS(Q) = B, A, E, D, A T Q

63 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING To find the correct results for the given query we find the subsequences of LPS(Q) inside LPS(T) Each subsequence represents a possible solution in the tree LPS(T) = A, C, B, C, C, B, A, C, A, E, E, E, D, A LPS(Q) = B, A, E, D, A 12 subsequences are found in total, while only 4 are correct T Q

64 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING To find the correct results for the given query we find the subsequences of LPS(Q) inside LPS(T) Each subsequence represents a possible solution in the tree LPS(T) = A, C, B, C, C, B, A, C, A, E, E, E, D, A LPS(Q) = B, A, E, D, A 12 subsequences are found in total, while only 4 are correct T Q

65 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING To find the correct results for the given query we find the subsequences of LPS(Q) inside LPS(T) Each subsequence represents a possible solution in the tree LPS(T) = A, C, B, C, C, B, A, C, A, E, E, E, D, A LPS(Q) = B, A, E, D, A 12 subsequences are found in total, while only 4 are correct T Q

66 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING To find the correct results for the given query we find the subsequences of LPS(Q) inside LPS(T) Each subsequence represents a possible solution in the tree LPS(T) = A, C, B, C, C, B, A, C, A, E, E, E, D, A LPS(Q) = B, A, E, D, A 12 subsequences are found in total, while only 4 are correct T Q

67 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING To find the correct results for the given query we find the subsequences of LPS(Q) inside LPS(T) Each subsequence represents a possible solution in the tree LPS(T) = A, C, B, C, C, B, A, C, A, E, E, E, D, A LPS(Q) = B, A, E, D, A 12 subsequences are found in total, while only 4 are correct T Q

68 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING To find the path in the tree that is represented by the sequence found while filtering we use the NPS(T) Recall that the edges can be retrieved using the index in the NPS(T) NPS(T) = 15, 3, 7, 6, 6, 7, 15, 9, 15, 13, 13, 13, 14, 15 LPS(T) = A, C, B, C, C, B, A, C, A, E, E, E, D, A T Q

69 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING To find the path in the tree that is represented by the sequence found while filtering we use the NPS(T) Recall that the edges can be retrieved using the index in the NPS(T) 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 NPS(T) = 15, 3, 7, 6, 6, 7, 15, 9, 15, 13, 13, 13, 14, 15 LPS(T) = A, C, B, C, C, B, A, C, A, E, E, E, D, A T Q

70 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING To find the path in the tree that is represented by the sequence found while filtering we use the NPS(T) Recall that the edges can be retrieved using the index in the NPS(T) 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 NPS(T) = 15, 3, 7, 6, 6, 7, 15, 9, 15, 13, 13, 13, 14, 15 LPS(T) = A, C, B, C, C, B, A, C, A, E, E, E, D, A T Q

71 Q UERYING : F ILTERING B Y S EQUENCE M ATCHING To find the path in the tree that is represented by the sequence found while filtering we use the NPS(T) Recall that the edges can be retrieved using the index in the NPS(T) 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 NPS(T) = 15, 3, 7, 6, 6, 7, 15, 9, 15, 13, 13, 13, 14, 15 LPS(T) = A, C, B, C, C, B, A, C, A, E, E, E, D, A T Q

72 Q UERYING : R EFINEMENT S TEPS Despite the filtering, some false positives are in the results. To find these false positives we have 3 refinement steps, namely: Refinement by connectedness Refinement by structure Refinement by matching leaf nodes

73 Q UERYING : F ALSE N EGATIVES A false negative can appear in the same case as in ViST index The subsequence filtering relies on the assumption that the query branches come in the “correct” order P F T N G Document P NF Query

74 Q UERYING : F ALSE N EGATIVES The solution proposed by Rao and Moon is to test the query in all possible permutations of the branches and then return the union as the answer of the query N branches  N! permutations Their main argument is that queries usually have a small number of branches

75 Q UERYING : F ALSE N EGATIVES The solution proposed by Rao and Moon is to test the query in all possible permutations of the branches and then return the union as the answer of the query N branches  N! permutations Their main argument is that queries usually have a small number of branches P NF D S P NFD S P NF D S … (three more permutations)

76 E XPERIMENTS

77 V I ST VS PRIX: E XPERIMENTS 1.8GHz Pentium IV processor 512 MB RAM running Solaris 8 40GB EIDE disk drive (store data and indexes) Compiled by GNU g++ compiler version 2.95.3 Buffer pool size: 2000 pages of size 8K

78 V I ST VS PRIX: E XPERIMENTS

79

80 DBLP dataset

81 V I ST VS PRIX: E XPERIMENTS SWISSPROT dataset

82 V I ST VS PRIX: E XPERIMENTS TREEBANK dataset

83 V I ST VS PRIX O(N 2 )

84 ?????? Q UESTIONS ?

85 T HANK YOU !!


Download ppt "S EQUENCE I NDEXING S CHEMES Roman Čížek Erasmus 2687, Nelly Vouzoukidou MET601."

Similar presentations


Ads by Google