1. ITCS 6265 IR & Web Mining ITCS 6265/8265: Advanced Topics in KDD --- Information Retrieval and Web Mining Lecture 1 Boolean retrieval UNC Charlotte, Fall 2009

2. ITCS 6265 IR & Web Mining Course administrivia  Web site: http://www.cs.uncc.edu/~wwu18/itcs6265/  Work/Grading:  Assignments20%  Project15%  Paper presentation 5%  Midterm30%  Final30%  Required textbook:  Introduction to Information Retrieval, by Manning et al. 2

3. ITCS 6265 IR & Web Mining Information Retrieval  Information Retrieval (IR) is finding material (usually documents) of an unstructured nature (usually text) that satisfies an information need from within large collections (usually stored on computers).  Unstructured data: structure & semantics of data are implicit, not specified to ease computer processing  Structured data, e.g., records in relational databases 3

4. ITCS 6265 IR & Web Mining Unstructured data in 1680  Which plays of Shakespeare contain the words Brutus AND Caesar but NOT Calpurnia?  One could grep all of Shakespeare’s plays for Brutus and Caesar, then strip out lines containing Calpurnia?  Slow (for large corpora, e.g., billions or trillions of words)  Other operations (e.g., find the word Romans near countrymen) not feasible  Ranked retrieval (best documents to return)  Later lectures  Need index! 4 Sec. 1.1

5. ITCS 6265 IR & Web Mining Take one: Index in term-document incidence matrix 1 if play contains word, 0 otherwise Brutus AND Caesar but NOT Calpurnia Sec. 1.1 Documents (plays) Words (~32k) …

6. ITCS 6265 IR & Web Mining Term vectors  So we have a 0/1 vector for each term.  To answer query: take the vectors for Brutus, Caesar and Calpurnia (complemented)  bitwise AND.  110100 AND 110111 AND 101111 = 100100. 6 Sec. 1.1

7. ITCS 6265 IR & Web Mining Answers to query  Antony and Cleopatra (Act III, Scene ii) Agrippa [Aside to DOMITIUS ENOBARBUS]: Why, Enobarbus, When Antony found Julius Caesar dead, He cried almost to roaring; and he wept When at Philippi he found Brutus slain.  Hamlet (Act III, Scene ii) Lord Polonius: I did enact Julius Caesar: I was killed i' the Capitol; Brutus killed me. 7 Sec. 1.1

8. ITCS 6265 IR & Web Mining Basic assumptions of Information Retrieval  Collection: Fixed set of documents  Goal: Retrieve documents with information that is relevant to user’s information need and helps him complete a task  Note: information need is topic of interest, different from query --- which are formulated by users based on their information need, & then submitted to retrieval systems to find relevant documents  But many things might go wrong in this process … 8 Sec. 1.1

9. ITCS 6265 IR & Web Mining The classic search model Corpus TASK Info Need Query Verbal form Results SEARCH ENGINE Query Refinement Get rid of mice in a politically correct way Info about removing mice without killing them How do I trap mice alive? mouse trap Mis-conceptionMis-translationMis-formulation

10. ITCS 6265 IR & Web Mining How good are the retrieved docs?  Precision : Fraction of retrieved docs that are relevant to user’s information need  Recall : Fraction of relevant docs in collection that are retrieved  More precise definitions and measurements to follow in later lectures 10 Sec. 1.1

11. ITCS 6265 IR & Web Mining Can matrix scale to larger collection?  Shakespeare’ collective works: ~36 plays  Now consider N = 1 million documents, each with about 1000 words. Note that this is still several orders of magnitude smaller than the size of Web!  Avg 6 bytes/word including spaces/punctuation  6GB (= 1M * 1000 * 6) of data in the documents.  Say there are M = 500K distinct terms among these. 11

12. ITCS 6265 IR & Web Mining Can’t build the matrix  500K x 1M matrix has half-a-trillion 0’s and 1’s.  But it has no more than one billion 1’s.  matrix is extremely sparse.  i.e., has a lot of zeros  What would be a better representation?  We only record the 1 positions. 12 Why?

13. ITCS 6265 IR & Web Mining Take two: Inverted index  For each term T, we must store a list of all documents that contain T.  Do we use an array or a list for this? 13 Brutus Calpurnia Caesar 12358132134 248163264128 1316 What happens if the word Caesar is added to document 14? Sec. 1.2 Sorted by docID (more later on why).

14. ITCS 6265 IR & Web Mining Inverted index  Linked lists generally preferred to arrays  Dynamic space allocation  Insertion of terms into documents easy  But have space overhead due to pointers 14 Brutus Calpurnia Caesar 248163264128 2358132134 1316 1 Dictionary Postings lists Posting Sec. 1.2

15. ITCS 6265 IR & Web Mining Tokenizer Token stream. Friends RomansCountrymen Inverted index construction Linguistic modules Normalized tokens (index terms). friend romancountryman Indexer Inverted index. friend roman countryman 24 2 13 16 1 More on these later. Documents to be indexed. Friends, Romans, countrymen. Sec. 1.2

16. ITCS 6265 IR & Web Mining Indexer steps  Sequence of (Normalized token, Document ID) pairs. I did enact Julius Caesar I was killed i' the Capitol; Brutus killed me. Doc 1 So let it be with Caesar. The noble Brutus hath told you Caesar was ambitious Doc 2 Sec. 1.2

17. ITCS 6265 IR & Web Mining Sorting  Sort by terms & then by doc ID’s Core indexing step. Sec. 1.2

18. ITCS 6265 IR & Web Mining Merging  Multiple term entries in the same document are merged.  Frequency information is added. Why frequency? Will discuss later. Sec. 1.2

19. ITCS 6265 IR & Web Mining Grouping (by term) & splitting (into dictionary and postings lists) Sec. 1.2

20. ITCS 6265 IR & Web Mining  Where do we pay in storage? 20 Pointers Terms Will quantify the storage, later. Sec. 1.2

21. ITCS 6265 IR & Web Mining The index we just built  How do we process a query?  Later - what kinds of queries can we process? 21 Today’s focus Sec. 1.3

22. ITCS 6265 IR & Web Mining Query processing: AND  Consider processing the query: Brutus AND Caesar  Locate Brutus in the Dictionary;  Retrieve its postings.  Locate Caesar in the Dictionary;  Retrieve its postings.  Intersect (sometimes called “merge”) the two postings: 22 128 34 248163264123581321 Brutus Caesar Sec. 1.3

23. ITCS 6265 IR & Web Mining The Intersect/Merge algorithm

24. ITCS 6265 IR & Web Mining The complexity of merge  Walk through the two postings simultaneously, in time linear in the total number of postings entries 24 34 12824816 3264 12 3 581321 128 34 248163264123581321 Brutus Caesar 2 8 If the list lengths are x and y, the merge takes O(x+y) operations. Crucial: postings sorted by docID. Sec. 1.3

25. ITCS 6265 IR & Web Mining Boolean queries: Exact match  The Boolean Retrieval model is being able to ask a query that is a Boolean expression:  Boolean Queries are queries using AND, OR and NOT to join query terms  Views each document as a set of words  Is precise: document matches condition or not.  Primary commercial retrieval tool for 3 decades.  Professional searchers (e.g., lawyers) still like Boolean queries:  You know exactly what you’re getting. 25 Sec. 1.3

26. ITCS 6265 IR & Web Mining Example: WestLaw http://www.westlaw.com/ http://library.uncc.edu/electronic/display.php?resource=434 (accessible to uncc) http://www.westlaw.com/ http://library.uncc.edu/electronic/display.php?resource=434  Largest commercial (paying subscribers) legal search service (started 1975; ranking added 1992)  Tens of terabytes of data; 700,000 users  Majority of users still use boolean queries  Example:  Information need: What is the statute of limitations in cases involving the federal tort claims act?  Query: LIMIT! /3 STATUTE ACT /S FEDERAL /2 TORT /3 CLAIM  /3 = within 3 words, /S = in same sentence 26 Sec. 1.4

27. ITCS 6265 IR & Web Mining Example: WestLaw http://www.westlaw.com/ http://www.westlaw.com/  Another example:  Information need: Requirements for disabled people to be able to access a workplace  Query: disabl! /p access! /s work-site work-place (employment /3 place)  Note that SPACE is disjunction, not conjunction!  Long, precise queries; proximity operators; incrementally developed; not like web search  Professional searchers often like Boolean search:  Precision, transparency and control  But that doesn’t mean they actually work better…. Sec. 1.4

28. ITCS 6265 IR & Web Mining Boolean queries: More general merges  Exercise: Adapt the merge for the queries: Brutus AND NOT Caesar Brutus OR NOT Caesar Can we still run through the merge in time O(x+y)? What can we achieve? 28 Sec. 1.3

29. ITCS 6265 IR & Web Mining Merging What about an arbitrary Boolean formula? (Brutus OR Caesar) AND NOT (Antony OR Cleopatra)  Can we always merge in “linear” time?  Linear in what?  Can we do better? 29 Sec. 1.3

30. ITCS 6265 IR & Web Mining Query optimization  What is the best order for query processing?  Consider a query that is an AND of t terms.  For each of the t terms, get its postings, then AND them together. Brutus Calpurnia Caesar 12358162134 248163264128 1316 Query: Brutus AND Calpurnia AND Caesar 30 Sec. 1.3

31. ITCS 6265 IR & Web Mining Query optimization example  Process in order of increasing freq:  start with smallest set, then keep cutting further. 31 Brutus Calpurnia Caesar 12358132134 248163264128 1316 This is why we kept freq in dictionary Execute the query as (Caesar AND Brutus) AND Calpurnia. Sec. 1.3

32. ITCS 6265 IR & Web Mining More general optimization  e.g., (madding OR crowd) AND (ignoble OR strife)  Get freq’s for all terms.  Estimate the size of each OR by the sum of its freq’s (conservative).  Process in increasing order of OR sizes. 32 Sec. 1.3

33. ITCS 6265 IR & Web Mining Exercise  Recommend a query processing order for 33 (tangerine OR trees) AND (marmalade OR skies) AND (kaleidoscope OR eyes)

34. ITCS 6265 IR & Web Mining What is ahead? Part I: IR & Search Engines  Positional index: phrase/proximity queries  Tolerant retrieval: wildcard, spelling correction, etc.  Efficient index construction  Index compression  Ranked retrieval: term weighting, vector space model  Evaluation of IR system  Relevance feedback & query expansion  Web search, crawling, and indexing 34

35. ITCS 6265 IR & Web Mining Part II: Text mining techniques  Classification: given a set of topics, plus a new doc D, decide which topic(s) D belongs to.  Naïve Bayes classifier, vector space classifier, support vector machine  Clustering: given a set of docs, group them into clusters based on their contents.  flat, hierarchical clustering, latent semantic indexing 35

36. ITCS 6265 IR & Web Mining Other text mining tasks (if time permits)  Entity extraction  Summarization  Opinion retrieval  Sentiment analysis  … 36

37. ITCS 6265 IR & Web Mining Part III: Web data mining  Content mining  Web data extraction, web data integration, etc.  Linkage analysis & structure mining  Authority, hub, pagerank, community discovery, etc  Usage mining  Web/application server log, user browsing history, etc.