Presentation is loading. Please wait.

Presentation is loading. Please wait.

Efficient Indexing of Shared Content in IR Systems Andrei Broder, Nadav Eiron, Marcus Fontoura, Michael Herscovici, Ronny Lempel, John McPherson, Eugene.

Similar presentations


Presentation on theme: "Efficient Indexing of Shared Content in IR Systems Andrei Broder, Nadav Eiron, Marcus Fontoura, Michael Herscovici, Ronny Lempel, John McPherson, Eugene."— Presentation transcript:

1 Efficient Indexing of Shared Content in IR Systems Andrei Broder, Nadav Eiron, Marcus Fontoura, Michael Herscovici, Ronny Lempel, John McPherson, Eugene Shekita, Runping Qi

2 Motivation IR systems typically use inverted indices to facilitate efficient retrieval Web, email, news, and other data contains significant amount of duplicated or shared content Indexing duplicate content is expensive

3 Scope of Work We assume duplicate or common content is already identified in the corpus We concern ourselves only with the efficient indexing of such content

4 Types of Shared Content Web duplicates: Very common – on the order of 40% of all pages Email/news threads: Whole messages are often quoted Attachments are duplicated Identical messages in multiple mailboxes

5 Some Statistics IBM Intranet has about 40% duplicate content. Internet crawls reveal similar statistics In the Enron email dataset, 61% of messages are in threads. 31% quote other messages verbatim

6 Na ï ve Solution 1 : Index Everything Pros: Simple to implement Semantics are preserved Cons: Index size blows up Performance penalty (big index + post filtering)

7 Na ï ve Solution 2: Index Just One Copy Pros: Best performance Not too difficult to implement Cons: Only applies to the duplicates scenario Semantics are changed, and relevant results may not be returned for a query

8 The Web Duplicate Case: Meta Data Vs. Content Removal of web duplicates changes the semantics of the query text http:// almaden.ibm.com /... text http:// watson.ibm.com /... Query: text url:watson

9 Our Solution Content is split to shared and private parts Shared content is indexed only once Private content (such as metadata in the Web duplicates case) is indexed for each document Index provides virtual cursors that simulate having all content indexed

10 Advantages Index size, build time, and query efficiency Precise semantics No need for post-filtering

11 Inverted Indices Index is sorted by term For each term, a sorted list of documents in which it appears is maintained (postings list) Each occurrence (posting) contains additional payload T 1 :, … T 2 :, …

12 Document Sharing Model Each document is partitioned into private and shared content. The two types are differentiated by posting payload Documents exist in a tree – shared content is shared with all descendents Document IDs (and hence index order) are dictated by a DFS traversal of document trees

13 The Document Tree Content is shared from ancestor to descendants: 1 2 3 4 56

14 Example: docid = 1: From: andrei To: ronny, marcus did you read it? docid = 2: From: ronny To: marcus did you, marcus? docid = 3: From: marcus To: ronny not yet! andrei: did:, it: marcus:,,, not: read: ronny:,, yet: you:, DocumentsInverted index posting lists 1 2 3 4 56

15 Querying Inverted Indexes Queries contain mandatory terms, forbidden terms, and optional terms (such as +term1 – term2) Typically a zigzag algorithm is used Uses cursors on postings list. Cursors support two operations: next() – Moves to the next posting fwdBeyond(d) – Moves to the first posting for a document with id >= d

16 Top Level Query Algorithm 1. while (more results required) { 2. Invoke zigzag algorithm 3. Forward optional term cursors 4. Score document 5. Advance required/forbidden cursors 6. } In our solution, this algorithm, uses virtual cursors

17 Additional Information In The Index Tree information is encoded by two attributes for each document: root(d) – The docid for the document at the root of the tree containing d lastDescendent(d) – The highest-numbered document that is a descendent of d

18 Physical Cursor Addition physicalCursor::fwdShare(d) 1. while (this.docid<=d and this.docid does not share content with d) { 2. r=root(d); 3. l=lastDescendant(this.docid); 4. if (this.docid<r) { 5. this.fwdBeyond(r); 6. } else if (l<d) { 7. this.fwdBeyond(l+1); 8. } else this.next(); 9. }

19 fwdShared(d) example: 1 2 34 5 6 7 8 910 p p p s s fwdShared(10)fwdBeyond(root(10))Next()fwdBeyond(lastDescendent(6)+1) T:,,,,

20 Virtual Cursors Two types of cursors: Regular (positive) virtual cursors. These behave as if all shared content was indexed for all documents that contain it Negated virtual cursors, represent the complement of the postings list (used for forbidden terms) Implemented on top of a physical cursor

21 Virtual Cursor Methods VirtualCursor::next() 1. l=lastDescendant(C p.docid) 2. if (C p.payload == shared and this.docid<l) 3. this.docid++; 4. else { 5. C p.next(); 6. this.docid=C p.docid; 7. } VirtualCursor::fwdBeyond(d) 1. if (this.docid>=d) 2. return; 3. C p.fwdShare(d); 4. this.docid = max(C p.docid,d);

22 Virtual Positive Cursors Maintain a physical and logical positions. Support next() and fwdBeyond(d) 1 2 34 5 6 7 8 910 p p p s s next()fwdBeyond(10)

23 Virtual Negative Cursors Support next() and fwdBeyond(d). Physical cursor ahead of logical cursor. 1 2 34 5 6 7 8 910 p p p s next()fwdBeyond(7) p

24 Web Duplicates Application Trees are flat, with the masters at the root. Leaves only have private content: docid = 1 root = 1 lastDescendant = 4 docid = 2 root = 1 lastDescendant = 2 docid = 3 root = 1 lastDescendant = 3 docid = 4 root = 1 lastDescendant = 4 S1S1 P1P1 P2P2 P3P3 P4P4 docid = 6 root = 5 lastDescendant = 6 S5S5 P5P5 P6P6

25 Build Performance Evaluation Subsets of IBM Intranet (36-44% dups): # docsIS1 (GB)IS2 (GB)Space saved IT1 (s)IT2 (s)Speedup 500K2.53.631%54078031% 1000K5.17.431%1020144029% 1500K7.111.036%1500234036% 2000K8.813.032%1800294039% 2500K11.016.031%2160354039%

26 Runtime Performance: Single Terms Queries

27 Runtime Performance: Two Term Queries


Download ppt "Efficient Indexing of Shared Content in IR Systems Andrei Broder, Nadav Eiron, Marcus Fontoura, Michael Herscovici, Ronny Lempel, John McPherson, Eugene."

Similar presentations


Ads by Google