Administrivia

1. Administrivia • Project?

2. Overview • History • Networking • Crawlers • Information Retrieval ?

3. Outline of IR topics • Background • Definitions, etc. • The Problem • 100,000+ pages • The Solution • Ranking docs • Vector space • Extensions • Relevance feedback, • clustering, • query expansion, etc.

4. Information retrieval Traditional Model Given a set of documents A query expressed as a set of keywords Return A ranked set of documents most relevant to the query Evaluation: Precision: Fraction of returned documents that are relevant Recall: Fraction of relevant documents that are returned Efficiency Web-induced headaches Scale (billions of documents) Hypertext (inter-document connections) Consequently Ranking that takes link structure into account Authority/Hub Indexing and Retrieval algorithms that are ultra fast

5. What is Information Retrieval • Given a large repository of documents, … • How do I get at the ones that I want • Examples: Lexus/Nexus, Medical reports, AltaVista • Keyword based [can’t handle synonymy, polysemy] • Different from databases • Unstructured (or semi-structured) data • Information is (typically) text • Requests are (typically) word-based In principle, this requires NLP! --NLP too hard as yet --IR tries to get by with syntactic methods

6. What is IR continued • IR = • representation, • storage, • organization of, and • access to • Focus is on the user information need • User information need: • Find all docs containing information on college tennis teams which: (1) are maintained by a USA university and (2) participate in the NCAA tournament. • Emphasis is: retrieval of information (not data) • } • information items

7. Information vs. Data • Data retrieval • which docs contain a set of keywords? • Well defined semantics • a single erroneous object implies failure! • Information retrieval • information about a subject or topic • semantics is frequently loose • small errors are tolerated • IR system: • interpret contents of information items • generate a ranking which reflects relevance • notion of relevance is most important

8. IR - Past and Present • IR at the center of the stage • IR in the last 20 years: • classification and categorization • systems and languages • user interfaces and visualization • The Web has renewed focus on IR • universal repository of knowledge • free (low cost) universal access • no central editorial board • many problems though: IR seen as key to finding the solutions!

9. Classic IR Models - Basic Concepts • Each document represented by a set of representative keywords or index terms • Query is seen as a “mini”document • An index term is a document word useful for remembering the document main themes • Usually, index terms are nouns because nouns have meaning by themselves • [However, search engines assume that all words are index terms (full text representation)]

10. Text User Interface 4, 10 user need Text Text Operations (stemming, noun phrase detection etc..) 6, 7 logical view logical view Query Operations (elaboration, relevance feedback DB Manager Module Indexing user feedback 5 8 inverted file query Searching (hash tables etc.) Index 8 retrieved docs Text Database Ranking (vector models ..) ranked docs 2 The Retrieval Process

11. structure Full text Index terms Generating keywords • Logical view of the documents Accents spacing Noun groups Manual indexing Docs stopwords stemming structure • Stop-word elimination • Noun phrase detection • Stemming • Generating index terms • Improving quality of terms. • Synonyms, co-occurence detection, latent semantic indexing..

12. Stop word elimination Stemming Example of Stemming and Stopword Elimination The number of Web pages on the World Wide Web was estimated to be over 800 million in 1999.

13. A quick glimpse at inverted files Dictionary Postings

14. Ranking • Is an ordering of the documents retrieved that (hopefully) reflects the relevance of the documents to the user query • Is based on fundamental premisses regarding the notion of relevance, such as: • common sets of index terms • sharing of weighted terms • likelihood of relevance • Each set of premisses leads to a distinct IR model

15. IR Models Algebraic Set Theoretic Generalized Vector Lat. Semantic Index Neural Networks Structured Models Fuzzy Extended Boolean Non-Overlapping Lists Proximal Nodes Classic Models Probabilistic boolean vector probabilistic Inference Network Belief Network Browsing Flat Structure Guided Hypertext U s e r T a s k Retrieval: Adhoc Filtering Browsing

16. How Measure Performance • Set of documents • Docs the user desires • User query

17. Measuring Performance Actual relevant docs • Precision • Proportion of selected items that are correct • Recall • Proportion of target items that were selected • Precision-Recall curve • Shows tradeoff tn fp tp fn System returned these Precision Recall

18. Precision / Recall Curves 11-point recall-precision curve Example: • Suppose for a given query10 documents are relevant. • Suppose when all documents are ranked in descending similarities, we have d1 d2d3 d4 d5d6 d7 d8 d9d10 d11d12d13 d14 d15 d16d17d18 d19d20 d21 d22 d23 d24d25d26 d27 d28d29 d30 d31 … precision 1.0 recall .1 .3

19. Precision Recall Curves… When evaluating the retrieval effectiveness of a text retrieval system or method, a large number of queries are used and their average 11-point recall-precision curve is plotted precision Method 1 Method 2 Method 3 recall

20. Methods 1 and 2 are better than method 3. • Method 1 is better than method 2 for high recalls.

21. IR Models Algebraic Set Theoretic Generalized Vector Lat. Semantic Index Neural Networks Structured Models Fuzzy Extended Boolean Non-Overlapping Lists Proximal Nodes Classic Models Probabilistic boolean vector probabilistic Inference Network Belief Network Browsing Flat Structure Guided Hypertext U s e r T a s k Retrieval: Adhoc Filtering Browsing

22. Documents as bags of words a: System and human system engineering testing of EPS b: A survey of user opinion of computer system response time c: The EPS user interface management system d: Human machine interface for ABC computer applications e: Relation of user perceived response time to error measurement f: The generation of random, binary, ordered trees g: The intersection graph of paths in trees h: Graph minors IV: Widths of trees and well-quasi-ordering i: Graph minors: A survey

23. Terminology: Term Weights • Not all terms are equally useful for representing the document contents • Less frequent terms allow identifying a narrower set of documents • The importance of the index terms is represented by weights associated to them • ki is an index term • dj is a document • t is the total number of docs • K = (k1, k2, …, kt) is the set of all index terms • wij >= 0 is a weight associated with (ki,dj) • wij = 0 indicates term missing from doc • vec(dj) = (w1j, w2j, …, wtj) is a weighted vector associated with the document dj (or query q)

24. The Boolean Model • Simple model based on set theory • Queries specified as boolean expressions • precise semantics • Terms are either present or absent. • Thus, wij  {0,1} • Consider • q = ka  (kb  kc) • dnf(q) = (1,1,1)  (1,1,0)  (1,0,0) • cc = (1,1,0) is a conjunctive component

25. The Boolean Model Illustrated Ka Kb (1,1,0) q = ka  (kb  kc) sim(q,dj) = 1 if  cc  dnf(q)  ki q, vec(dj)i = cci 0 otherwise (1,0,0) (1,1,1) Kc

26. Evaluation of Boolean Model

27. Drawbacks of the Boolean Model • Binary decision criteria • No notion of partial matching • No ranking or grading scale • Users must write Boolean expression • Awkward • Often too simplistic • Thus users get too few or too many documents

28. Thus ... The Vector Model • Use of binary weights is too limiting • [0, 1] term weights are used to compute • Degree of similarity between a query and documents • Allows ranking of results

29. The Vector Model Definitions • Documents/Queries modeled as bags of words • Represented as vectors over keyword space • vec(dj) = (w1j, w2j, ..., wtj) • vec(q) = (w1q, w2q, ..., wtq) • wij > 0 whenever ki  dj • wiq >= 0 associated with the pair (ki,q) • To each term ki is associated a unitary vector vec(i) • Unitary vectors vec(i) and vec(j) are assumed orthonormal • What does this mean? • Is this Reasonable?????? t unitary vectors vec(i) form • an orthonormal basis for a t-dimensional space • Each vector holds a place for every term in the collection • Therefore, most vectors are sparse

30. Vector Space Example a: System and human system engineering testing of EPS b: A survey of user opinion of computer system response time c: The EPS user interface management system d: Human machine interface for ABC computer applications e: Relation of user perceived response time to error measurement f: The generation of random, binary, ordered trees g: The intersection graph of paths in trees h: Graph minors IV: Widths of trees and well-quasi-ordering i: Graph minors: A survey Documents

31. Similarity Function The similarity or closeness of a document d = ( w1, …, wi, …, wn ) with respect to a query (or another document) q = ( q1, …, qi, …, qn ) is computed using a similarity (distance) function. Many similarity functions exist

32. Euclidian Distance • Given two document vectors d1 and d2

33. Dot Product Distance sim(q, d) = dot(q, d) = q1  w1 + … + qn wn Example: Suppose d = (0.2, 0, 0.3, 1) and q = (0.75, 0.75, 0, 1), then sim(q, d) = 0.15 + 0 + 0 + 1 = 1.15 Observations of the dot product function. • More terms in common => higher similarity • For terms appearing in both q and d, • those with higher weights contribute more to sim(q, d) • Favors long documents over short documents. • The computed similarities have no clear upper bound.

34. A normalized similarity metric j dj  q i • Sim(q,dj) = cos() = [vec(dj)  vec(q)] / |dj| * |q| = [ wij * wiq] / |dj| * |q| • Since wij > 0 and wiq > 0, 0 <= sim(q,dj) <=1 • A document is retrieved even if it matches the query terms only partially

35. Eucledian Cosine t1= database t2=SQL t3=index t4=regression t5=likelihood t6=linear Comparison of Eucledian and Cosine distance metrics

36. Answering Queries • Represent query as vector • Compute distances to all documents • Rank according to distance • Example • “database index” t1= database t2=SQL t3=index t4=regression t5=likelihood t6=linear Given Q={database, index} = {1,0,1,0,0,0}

37. Term Weights in the Vector Model • Sim(q,dj) = [ wij * wiq] / |dj| * |q| • How to compute the weights wij and wiq ?

38. Simple frequencies favor common words • E.g. Query: The Computer Tomography • A good weight must take into account two effects: • Intra-document contents (similarity) • tf factor, the term frequency within a document • Inter-document separation (dis-similarity) • idf factor, the inverse document frequency wij = tf(i,j) * idf(i)

39. TF-IDF • Let, • N be the total number of docs in the collection • ni be the number of docs which contain ki • freq(i,j) raw frequency of ki within dj • A normalized tf factor is given by • f(i,j) = freq(i,j) / max(freq(i,j)) • where the maximum is computed over all terms which occur within the document dj • The idf factor is computed as • idf(i) = log (N/ni) • the log is used to make the values of tf and idf comparable. • Can be interpreted as the amount of information associated with the term ki.

40. Document/Query Representation using TF-IDF • The best schemes use weights like • wij = f(i,j) * log(N/ni) • the strategy is called a tf-idfweighting scheme • For the query term weights, several possibilities: • wiq = (0.5 + [0.5 * freq(i,q) / max(freq(i,q)]) * log(N/ni) • Or use the IDF weights (to give preference to rare words) • Or let the user give weights to reflect her real preferences • Easier said than done... Users are often dunderheads.. • Help them with “relevance feedback” techniques.

41. t1= database t2=SQL t3=index t4=regression t5=likelihood t6=linear Note: In this case, the weights used in query were 1 for t1 and t3, and 0 for the rest. Given Q={database, index} = {1,0,1,0,0,0}

42. The Vector Model:Summary • The vector model with tf-idf weights is ~ best • Usually as good as the known ranking alternatives. • Simple and fast to compute. • Advantages: • term-weighting improves quality of the answer set • partial matching • cosine ranking formula sorts documents by query similarity • Disadvantages: • Assumes independence of index terms • Does not handle synonymy/polysemy • Query weighting may not reflect user relevance criteria.