Question? Leave a message!




How can we improve recall in search

How can we improve recall in search
Introduction to Information Retrieval Introduction to Information Retrieval Relevance Feedback Query Expansion 1Introduction to Information Retrieval Takeaway today  Interactive relevance feedback: improve initial retrieval results by telling the IR system which docs are relevant / nonrelevant  Best known relevance feedback method: Rocchio feedback  Query expansion: improve retrieval results by adding synonyms / related terms to the query  Sources for related terms: Manual thesauri, automatic thesauri, query logs 2 2Introduction to Information Retrieval Overview ❶ Motivation ❷ Relevance feedback: Basics ❸ Relevance feedback: Details ❹ Query expansion 3Introduction to Information Retrieval Outline ❶ Motivation ❷ Relevance feedback: Basics ❸ Relevance feedback: Details ❹ Query expansion 4Introduction to Information Retrieval How can we improve recall in search  Main topic today: two ways of improving recall: relevance feedback and query expansion  As an example consider query q: aircraft . . .  . . . and document d containing “plane”, but not containing “aircraft”  A simple IR system will not return d for q.  Even if d is the most relevant document for q  We want to change this:  Return relevant documents even if there is no term match with the (original) query 5 5Introduction to Information Retrieval Recall  Loose definition of recall in this lecture: “increasing the number of relevant documents returned to user”  This may actually decrease recall on some measures, e.g., when expanding “jaguar” with “panthera”  . . .which eliminates some relevant documents, but increases relevant documents returned on top pages 6 6Introduction to Information Retrieval Options for improving recall  Local: Do a “local”, ondemand analysis for a user query  Main local method: relevance feedback  Part 1  Global: Do a global analysis once (e.g., of collection) to produce thesaurus  Use thesaurus for query expansion  Part 2 7 7Introduction to Information Retrieval Google examples for query expansion  One that works well  ˜flights flight  One that doesn’t work so well  ˜hospitals hospital 8 8Introduction to Information Retrieval Outline ❶ Motivation ❷ Relevance feedback: Basics ❸ Relevance feedback: Details ❹ Query expansion 9Introduction to Information Retrieval Relevance feedback: Basic idea  The user issues a (short, simple) query.  The search engine returns a set of documents.  User marks some docs as relevant, some as nonrelevant.  Search engine computes a new representation of the information need. Hope: better than the initial query.  Search engine runs new query and returns new results.  New results have (hopefully) better recall. 10 10Introduction to Information Retrieval Relevance feedback  We can iterate this: several rounds of relevance feedback.  We will use the term ad hoc retrieval to refer to regular retrieval without relevance feedback.  We will now look at three different examples of relevance feedback that highlight different aspects of the process. 11 11Introduction to Information Retrieval Relevance feedback: Example 1 12 12Introduction to Information Retrieval Results for initial query 13 13Introduction to Information Retrieval User feedback: Select what is relevant 14 14Introduction to Information Retrieval Results after relevance feedback 15 15Introduction to Information Retrieval Vector space example: query “canine” (1) Source: Fernando Díaz 16 16Introduction to Information Retrieval Similarity of docs to query “canine” Source: Fernando Díaz 17 17Introduction to Information Retrieval User feedback: Select relevant documents Source: Fernando Díaz 18 18Introduction to Information Retrieval Results after relevance feedback Source: Fernando Díaz 19 19Introduction to Information Retrieval Example 3: A real (nonimage) example Initial query: new space satellite applications Results for initial query: (r = rank) r + 1 0.539 NASA Hasn’t Scrapped Imaging Spectrometer + 2 0.533 NASA Scratches Environment Gear From Satellite Plan 3 0.528 Science Panel Backs NASA Satellite Plan, But Urges Launches of Smaller Probes 4 0.526 A NASA Satellite Project Accomplishes Incredible Feat: Staying Within Budget 5 0.525 Scientist Who Exposed Global Warming Proposes Satellites for Climate Research 6 0.524 Report Provides Support for the Critics Of Using Big Satellites to Study Climate 7 0.516 Arianespace Receives Satellite Launch Pact From Telesat Canada + 8 0.509 Telecommunications Tale of Two Companies User then marks relevant documents with “+”. 20Introduction to Information Retrieval Expanded query after relevance feedback 2.074 new 15.106 space 30.816 satellite 5.660 application 5.991 nasa 5.196 eos 4.196 launch 3.972 aster 3.516 instrument 3.446 arianespace Compare to original 3.004 bundespost 2.806 ss 2.790 rocket 2.053 scientist 2.003 broadcast 1.172 earth 0.836 oil 0.646 measure query: new space satellite applications 21 21Introduction to Information Retrieval Results for expanded query r 1 0.513 NASA Scratches Environment Gear From Satellite Plan 2 0.500 NASA Hasn’t Scrapped Imaging Spectrometer 3 0.493 When the Pentagon Launches a Secret Satellite, Space Sleuths Do Some Spy Work of Their Own 4 0.493 NASA Uses ‘Warm’ Superconductors For Fast Circuit 5 0.492 Telecommunications Tale of Two Companies 6 0.491 Soviets May Adapt Parts of SS20 Missile For Commercial Use 7 0.490 Gaping Gap: Pentagon Lags in Race To Match the Soviets In Rocket Launchers 8 0.490 Rescue of Satellite By Space Agency To Cost 90 Million 22Introduction to Information Retrieval Outline ❶ Motivation ❷ Relevance feedback: Basics ❸ Relevance feedback: Details ❹ Query expansion 23Introduction to Information Retrieval Key concept for relevance feedback: Centroid  The centroid is the center of mass of a set of points.  Recall that we represent documents as points in a high dimensional space.  Thus: we can compute centroids of documents.  Definition: where D is a set of documents and is the vector we use to represent document d. 24 24Introduction to Information Retrieval Centroid: Example 25 25Introduction to Information Retrieval Rocchio’ algorithm  The Rocchio’ algorithm implements relevance feedback in the vector space model.  Rocchio’ chooses the query that maximizes D : set of relevant docs; D : set of nonrelevant docs r nr  Intent: qopt is the vector that separates relevant and nonrelevant docs maximally.  Making some additional assumptions, we can rewrite as: 26 26Introduction to Information Retrieval Rocchio’ algorithm  The optimal query vector is:  We move the centroid of the relevant documents by the difference between the two centroids. 27 27Introduction to Information Retrieval Exercise: Compute Rocchio’ vector circles: relevant documents, Xs: nonrelevant documents 28 28Introduction to Information Retrieval Rocchio’ illustrated : centroid of relevant documents 29 29Introduction to Information Retrieval Rocchio’ illustrated does not separate relevant / nonrelevant. 30 30Introduction to Information Retrieval Rocchio’ illustrated centroid of nonrelevant documents. 31 31Introduction to Information Retrieval Rocchio’ illustrated 32 32Introduction to Information Retrieval Rocchio’ illustrated difference vector 33 33Introduction to Information Retrieval Rocchio’ illustrated Add difference vector to … 34 34Introduction to Information Retrieval Rocchio’ illustrated … to get 35 35Introduction to Information Retrieval Rocchio’ illustrated separates relevant / nonrelevant perfectly. 36 36Introduction to Information Retrieval Rocchio’ illustrated separates relevant / nonrelevant perfectly. 37 37Introduction to Information Retrieval Terminology  We use the name Rocchio’ for the theoretically better motivated original version of Rocchio.  The implementation that is actually used in most cases is the SMART implementation – we use the name Rocchio (without prime) for that. 38 38Introduction to Information Retrieval Rocchio 1971 algorithm (SMART) Used in practice: q : modified query vector; q : original query vector; D and m 0 r D : sets of known relevant and nonrelevant documents nr respectively; α, β, and γ: weights  New query moves towards relevant documents and away from nonrelevant documents.  Tradeoff α vs. β/γ: If we have a lot of judged documents, we want a higher β/γ.  Set negative term weights to 0.  “Negative weight” for a term doesn’t make sense in the 39 39 vector space model.Introduction to Information Retrieval Positive vs. negative relevance feedback  Positive feedback is more valuable than negative feedback.  For example, set β = 0.75, γ = 0.25 to give higher weight to positive feedback.  Many systems only allow positive feedback. 40 40Introduction to Information Retrieval Relevance feedback: Assumptions  When can relevance feedback enhance recall  Assumption A1: The user knows the terms in the collection well enough for an initial query.  Assumption A2: Relevant documents contain similar terms (so I can “hop” from one relevant document to a different one when giving relevance feedback). 41 41Introduction to Information Retrieval Violation of A1  Assumption A1: The user knows the terms in the collection well enough for an initial query.  Violation: Mismatch of searcher’s vocabulary and collection vocabulary  Example: cosmonaut / astronaut 42 42Introduction to Information Retrieval Violation of A2  Assumption A2: Relevant documents are similar.  Example for violation: contradictory government policies  Several unrelated “prototypes”  Subsidies for tobacco farmers vs. antismoking campaigns  Aid for developing countries vs. high tariffs on imports from developing countries  Relevance feedback on tobacco docs will not help with finding docs on developing countries. 43 43Introduction to Information Retrieval Relevance feedback: Evaluation  Pick one of the evaluation measures from last lecture, e.g., precision in top 10: P10  Compute P10 for original query q 0  Compute P10 for modified relevance feedback query q1  In most cases: q is spectacularly better than q 1 0  Is this a fair evaluation 44 44Introduction to Information Retrieval Relevance feedback: Evaluation  Fair evaluation must be on “residual” collection: docs not yet judged by user.  Studies have shown that relevance feedback is successful when evaluated this way.  Empirically, one round of relevance feedback is often very useful. Two rounds are marginally useful. 45 45Introduction to Information Retrieval Evaluation: Caveat  True evaluation of usefulness must compare to other methods taking the same amount of time.  Alternative to relevance feedback: User revises and resubmits query.  Users may prefer revision/resubmission to having to judge relevance of documents.  There is no clear evidence that relevance feedback is the “best use” of the user’s time. 46 46Introduction to Information Retrieval Exercise  Do search engines use relevance feedback  Why 47 47Introduction to Information Retrieval Relevance feedback: Problems  Relevance feedback is expensive.  Relevance feedback creates long modified queries.  Long queries are expensive to process.  Users are reluctant to provide explicit feedback.  It’s often hard to understand why a particular document was retrieved after applying relevance feedback.  The search engine Excite had full relevance feedback at one point, but abandoned it later. 48 48Introduction to Information Retrieval Pseudorelevance feedback  Pseudorelevance feedback automates the “manual” part of true relevance feedback.  Pseudorelevance algorithm:  Retrieve a ranked list of hits for the user’s query  Assume that the top k documents are relevant.  Do relevance feedback (e.g., Rocchio)  Works very well on average  But can go horribly wrong for some queries.  Several iterations can cause query drift. 49 49Introduction to Information Retrieval Pseudorelevance feedback at TREC4  Cornell SMART system  Results show number of relevant documents out of top 100 for 50 queries (so total number of documents is 5000): method number of relevant documents lnc.ltc 3210 lnc.ltcPsRF 3634 Lnu.ltu 3709 Lnu.ltuPsRF 4350  Results contrast two length normalization schemes (L vs. l) and pseudorelevance feedback (PsRF).  The pseudorelevance feedback method used added only 20 terms to the query. (Rocchio will add many more.)  This demonstrates that pseudorelevance feedback is effective on average. 50 50Introduction to Information Retrieval Outline ❶ Motivation ❷ Relevance feedback: Basics ❸ Relevance feedback: Details ❹ Query expansion 51Introduction to Information Retrieval Query expansion  Query expansion is another method for increasing recall.  We use “global query expansion” to refer to “global methods for query reformulation”.  In global query expansion, the query is modified based on some global resource, i.e. a resource that is not query dependent.  Main information we use: (near)synonymy  A publication or database that collects (near)synonyms is called a thesaurus.  We will look at two types of thesauri: manually created and automatically created. 52 52Introduction to Information Retrieval Query expansion: Example 53 53Introduction to Information Retrieval Types of user feedback  User gives feedback on documents.  More common in relevance feedback  User gives feedback on words or phrases.  More common in query expansion 54 54Introduction to Information Retrieval Types of query expansion  Manual thesaurus (maintained by editors, e.g., PubMed)  Automatically derived thesaurus (e.g., based on co occurrence statistics)  Queryequivalence based on query log mining (common on the web as in the “palm” example) 55 55Introduction to Information Retrieval Thesaurusbased query expansion  For each term t in the query, expand the query with words the thesaurus lists as semantically related with t.  Example from earlier: HOSPITAL → MEDICAL  Generally increases recall  May significantly decrease precision, particularly with ambiguous terms  INTEREST RATE → INTEREST RATE FASCINATE  Widely used in specialized search engines for science and engineering  It’s very expensive to create a manual thesaurus and to maintain it over time.  A manual thesaurus has an effect roughly equivalent to annotation with a controlled vocabulary. 56 56Introduction to Information Retrieval Example for manual thesaurus: PubMed 57 57Introduction to Information Retrieval Automatic thesaurus generation  Attempt to generate a thesaurus automatically by analyzing the distribution of words in documents  Fundamental notion: similarity between two words  Definition 1: Two words are similar if they cooccur with similar words.  “car” ≈ “motorcycle” because both occur with “road”, “gas” and “license”, so they must be similar.  Definition 2: Two words are similar if they occur in a given grammatical relation with the same words.  You can harvest, peel, eat, prepare, etc. apples and pears, so apples and pears must be similar.  Cooccurrence is more robust, grammatical relations are more accurate. 58 58Introduction to Information Retrieval Cooccurencebased thesaurus: Examples Word Nearest neighbors absolutely absurd whatsoever totally exactly nothing bottomed dip copper drops topped slide trimmed captivating shimmer stunningly superbly plucky witty doghouse dog porch crawling beside downstairs makeup repellent lotion glossy sunscreen skin gel mediating reconciliation negotiate case conciliation keeping hoping bring wiping could some would lithographs drawings Picasso Dali sculptures Gauguin pathogens toxins bacteria organisms bacterial parasite senses grasp psyche truly clumsy naive innate WordSpace demo on web 59 59Introduction to Information Retrieval Query expansion at search engines  Main source of query expansion at search engines: query logs  Example 1: After issuing the query herbs, users frequently search for herbal remedies.  → “herbal remedies” is potential expansion of “herb”.  Example 2: Users searching for flower pix frequently click on the URL photobucket.com/flower. Users searching for flower clipart frequently click on the same URL.  → “flower clipart” and “flower pix” are potential expansions of each other. 60 60Introduction to Information Retrieval Takeaway today  Interactive relevance feedback: improve initial retrieval results by telling the IR system which docs are relevant / nonrelevant  Best known relevance feedback method: Rocchio feedback  Query expansion: improve retrieval results by adding synonyms / related terms to the query  Sources for related terms: Manual thesauri, automatic thesauri, query logs 61 61Introduction to Information Retrieval Resources  Chapter 9 of IIR  Resources at http://ifnlp.org/ir  Salton and Buckley 1990 (original relevance feedback paper)  Spink, Jansen, Ozmultu 2000: Relevance feedback at Excite  Schütze 1998: Automatic word sense discrimination (describes a simple method for automatic thesuarus generation) 62 62
Website URL
Comment