1. © Johan Bos November 2005 Carol Beer (Little Britain)

2. © Johan Bos November 2005 Computer says “no”

3. © Johan Bos November 2005 Question Answering Lecture 1 (Today): Introduction; History of QA; Architecture of a QA system; Evaluation. Lecture 2 (Friday): Question Classification; NLP techniques for question analysis; POS-tagging; Parsing; Semantic analysis; WordNet. Lecture 3 (Next Monday): Retrieving Answers; Document pre-processing; Tokenisation; Stemming; Lemmatisation; Named Entity Recognition; Anaphora Resolution; Matching; Use of knowledge resources; Reranking; Sanity checking.

4. © Johan Bos November 2005 What is Question Answering? ?

5. © Johan Bos November 2005 Information Pinpointing Information required: Average number of car accidents per year in Sweden. Two ways of getting this information: -Ask Google or a similar search engine (good luck!) -Ask a QA system the question: What’s the rate of car accidents in Sweden?

6. © Johan Bos November 2005 QA vs IR Traditional method for information access: IR (Information Retrieval) –Think of IR as finding the “right book in a library” –Think of QA as a “librarian giving you the book and opening it on the page with the information you’re looking for”

7. © Johan Bos November 2005 QA vs IE Traditional method for information access: IE (Information Extraction) –Think of IE as finding answers to a pre- defined question (i.e., a template) –Think of QA as asking any question you like

8. © Johan Bos November 2005 What is Question Answering? Questions in natural language, not queries! Answers, not documents!

9. © Johan Bos November 2005 Why do we need QA? Information overload problem Accessing information using traditional methods such as IR and IE are limited QA increasingly important because: –Size of available information grows –There is duplicate information –There is false information –More and more “computer illiterates” accessing electronically stored information

10. © Johan Bos November 2005 Information Avalanche Available information is growing*: –1999: 250MB pp for each person on earth –2002: 800MB pp for each person on earth People want specific information * source: M.de Rijke 2005

11. © Johan Bos November 2005 People ask Questions* * source: M.de Rijke 2005

12. © Johan Bos November 2005 Why is QA hard? (1/3) Questions are expressed in natural language (such as English or Italian) Unlike formal languages, natural languages allow a great deal of flexibility Example: –What is the population of Rome? –How many people live in Rome? –What’s the size of Rome? –How many inhabitants does Rome have?

13. © Johan Bos November 2005 Why is QA hard? (2/3) Answers are expressed in natural language (such as English or Italian) Unlike formal languages, natural languages allow a great deal of flexibility Example: …is estimated at 2.5 million residents… … current population of Rome is 2817000… …Rome housed over 1 million inhabitants…

14. © Johan Bos November 2005 Why is QA hard? (3/3) Answers could be spread across different documents Examples: –Which European countries produce wine? [Document A contains information about Italy, and document B about France] –What does Bill Clinton’s wife do for a living? [Document A explains that Bill Clinton’s wife is Hillary Clinton, and Document B tells us that she’s a politician]

15. © Johan Bos November 2005 History of QA (de Rijke & Webber 2003) QA is by no means a new area! Simmons (1965) reviews 15 implemented and working systems Many ingredients of today’s QA systems are rooted in these early approaches Database oriented systems, domain independent, as opposed to today’s systems that work on large sets of unstructured texts

16. © Johan Bos November 2005 Examples of early QA systems BASEBALL (Green et al. 1963) Answers English questions about scores, locations and dates of baseball games LUNAR (Woods 1977) Accesses chemical data on lunar material compiled during the Apollo missions PHLIQA1 (Scha et al. 1980) Answers short questions against a database of computer installations in Europe

17. © Johan Bos November 2005 Recent work in QA Since the 1990s research in QA has by and large focused on open-domain applications Recently interest in restricted-domain QA has increased, in particular in commercial applications –Banking, entertainment, etc.

18. © Johan Bos November 2005 Architecture of a QA system IR Question Analysis query Document Analysis Answer Extraction question answer-type question representation documents/passages passage representation corpus answers

19. © Johan Bos November 2005 Question Analysis Input: Natural Language Question Output: Expected Answer Type (Formal) Representation of Question Techniques used: Machine learning, parsing

20. © Johan Bos November 2005 Document Analysis Input: Documents or Passages Output: (Formal) Representation of Passages that might contain the answer Techniques used: Tokenisation, Named Entity Recognition, Parsing

21. © Johan Bos November 2005 Answer Retrieval Input: Expected Answer Type Question (formal representation) Passages (formal representation) Output: Ranked list of answers Techniques used: Matching, Re-ranking, Validation

22. © Johan Bos November 2005 Example Run IR Question Analysis query Document Analysis Answer Extraction question answer-type question representation documents/passages passage representation corpus answers

23. © Johan Bos November 2005 Example Run IR Question Analysis query Document Analysis Answer Extraction question answer-type question representation documents/passages passage representation corpus answers How long is the river Thames?

24. © Johan Bos November 2005 Example Run IR Question Analysis query Document Analysis Answer Extraction question answer-type question representation documents/passages passage representation corpus answers length river thames

25. © Johan Bos November 2005 Example Run IR Question Analysis query Document Analysis Answer Extraction question answer-type question representation documents/passages passage representation corpus answers MEASURE

26. © Johan Bos November 2005 Example Run IR Question Analysis query Document Analysis Answer Extraction question answer-type question representation documents/passages passage representation corpus answers Answer(x) & length(y,x) & river(y) & named(y,thames)

27. © Johan Bos November 2005 Example Run IR Question Analysis query Document Analysis Answer Extraction question answer-type question representation documents/passages passage representation corpus answers A: NYT199802-31 B: APW199805-12 C: NYT200011-07

28. © Johan Bos November 2005 Example Run IR Question Analysis query Document Analysis Answer Extraction question answer-type question representation documents/passages passage representation corpus answers A: 30(u) & mile(u) & length(v,u) & river(y) B: 60(z) & centimeter(z) & height(v,z) & dog(z) C: 230(u) & kilometer(u) & length(x,u) & river(x)

29. © Johan Bos November 2005 Example Run IR Question Analysis query Document Analysis Answer Extraction question answer-type question representation documents/passages passage representation corpus answers C: 230 kilometer A: 30 miles B: 60 centimeter

30. © Johan Bos November 2005 Evaluating QA systems International evaluation campaigns for QA systems (open domain QA): –TREC (Text Retrieval Conference) http://trec.nist.gov/ –CLEF (Cross Language Evaluation Forum) http://clef-qa.itc.it/ http://clef-qa.itc.it/ –NTCIR (NII Test Collection for IR Systems) http://www.slt.atr.jp/CLQA/ http://www.slt.atr.jp/CLQA/

31. © Johan Bos November 2005 TREC-QA (organised by NIST) Annual event, started in 1999 Difficulty of the QA task increased over the years: –1999: Answers in snippets, ranked list of answers; –2005: Exact answers, only one answer. Three types of questions: –Factoid questions –List questions –Definition questions

32. © Johan Bos November 2005 QA@CLEF CLEF is the “European edition” of TREC Monolingual (non-English) QA –Bulgarian (BG), German (DE), Spanish (ES), Finnish (FI), French (FR), Italian (IT), Dutch (NL), Portuguese (PT) Cross-Lingual QA –Questions posed in source language, answer searched in documents of target language –All combinations possible

33. © Johan Bos November 2005 Open-Domain QA QA at TREC is considered “Open-Domain” QA –Document collection is Acquint Corpus (over a million documents) –Questions can be about anything Restricted-Domain QA –Documents described a specific domain –Detailed questions –Less redundancy of answers!

34. © Johan Bos November 2005 TREC-type questions Factoid questions –Where is the Taj Mahal? List questions –What actors have played Tevye in `Fiddler on the Roof'? Definition/biographical questions –What is a golden parachute? –Who is Vlad the Impaler?

35. © Johan Bos November 2005 What is a correct answer? Example Factoid Question –When did Franz Kafka die? Possible Answers: –Kafka died in 1923. –Kafka died in 1924. –Kafka died on June 3, 1924 from complications related to Tuberculosis. –Ernest Watz was born June 3, 1924. –Kafka died on June 3, 1924.

36. © Johan Bos November 2005 What is a correct answer? Example Factoid Question –When did Franz Kafka die? Possible Answers: –Kafka died in 1923. –Kafka died in 1924. –Kafka died on June 3, 1924 from complications related to Tuberculosis. –Ernest Watz was born June 3, 1924. –Kafka died on June 3, 1924. Incorrect

37. © Johan Bos November 2005 What is a correct answer? Example Factoid Question –When did Franz Kafka die? Possible Answers: –Kafka died in 1923. –Kafka died in 1924. –Kafka died on June 3, 1924 from complications related to Tuberculosis. –Ernest Watz was born June 3, 1924. –Kafka died on June 3, 1924. Inexact (under-informative)

38. © Johan Bos November 2005 What is a correct answer? Example Question –When did Franz Kafka die? Possible Answers: –Kafka died in 1923. –Kafka died in 1924. –Kafka died on June 3, 1924 from complications related to Tuberculosis. –Ernest Watz was born June 3, 1924. –Kafka died on June 3, 1924. Inexact (over-informative)

39. © Johan Bos November 2005 What is a correct answer? Example Question –When did Franz Kafka die? Possible Answers: –Kafka died in 1923. –Kafka died in 1924. –Kafka died on June 3, 1924 from complications related to Tuberculosis. –Ernest Watz was born June 3, 1924. –Kafka died on June 3, 1924. Unsupported

40. © Johan Bos November 2005 What is a correct answer? Example Question –When did Franz Kafka die? Possible Answers: –Kafka died in 1923. –Kafka died in 1924. –Kafka died on June 3, 1924 from complications related to Tuberculosis. –Ernest Watz was born June 3, 1924. –Kafka died on June 3, 1924. Correct

41. © Johan Bos November 2005 Answer Accuracy # correct answers Answer Accuracy = --------------------------- # questions

42. © Johan Bos November 2005 Correct answers to list questions System A: France Italy Example List Question Which European countries produce wine? System B: Scotland France Germany Italy Spain Iceland Greece the Netherlands Japan Turkey Estonia

43. © Johan Bos November 2005 Evaluation metrics for list questions Precision (P): # answers judged correct & distinct P = ---------------------------------------------- # answers returned Recall (R): # answers judged correct & distinct R = ------------------------------------------------ # total answers F-Score (F): 2*P*R F = ------------ P+R

44. © Johan Bos November 2005 Correct answers to list questions System A: France Italy Example List Question Which European countries produce wine? System B: Scotland France Germany Italy Spain Iceland Greece the Netherlands Japan Turkey Estonia P = 1.00 R = 0.25 F = 0.40 P = 0.64 R = 0.88 F = 0.74

45. © Johan Bos November 2005 Other evaluation metrics System A: Ranked answers (Accuracy = 0.2) Q1Q2Q3Q4Q6Q7Q8Q9….Qn A1 WWCWCWWW….W A2 WWWWWWWW….W A3 WWWWWWWW….W A4 WWWWWWWW….W A5 WCWWWCWW….W System B: Ranked answers (Accuracy = 0.1) Q1Q2Q3Q4Q6Q7Q8Q9….Qn A1 WWWWCWWW….W A2 CWCWWCCW….C A3 WCWWWWWW….W A4 WWWCWWWW….W A5 WWWWWWWW….W

46. © Johan Bos November 2005 Mean Reciprocal Rank (MRR) Score for an individual question: –The reciprocal of the rank at which the first correct answer is returned –0 if no correct response is returned The score for a run: –Mean over the set of questions in the test

47. © Johan Bos November 2005 MRR in action System A: MRR = (.2+1+1+.2)/10 = 0.24 Q1Q2Q3Q4Q6Q7Q8Q9….Qn A1 WWCWCWWW….W A2 WWWWWWWW….W A3 WWWWWWWW….W A4 WWWWWWWW….W A5 WCWWWCWW….W System B: MRR = (.5+.33+.5+.25+1+.5+.5+.5)/10=0.42 Q1Q2Q3Q4Q6Q7Q8Q9….Qn A1 WWWWCWWW….W A2 CWCWWCCW….C A3 WCWWWWWW….W A4 WWWCWWWW….W A5 WWWWWWWW….W

48. © Johan Bos November 2005 Open-Domain Question Answering TREC QA Track –Factoid questions –List questions –Definition questions State-of-the-Art –Hard problem –Only few systems with good results

49. © Johan Bos November 2005 Friday QA Lecture 2: –Question Classification –NLP techniques for question analysis –POS-tagging –Parsing –Semantic analysis –Use of lexical resources such as WordNet

50. © Johan Bos November 2005 Question Classification (preview) How many islands does Italy have? When did Inter win the Scudetto? What are the colours of the Lithuanian flag? Where is St. Andrews located? Why does oil float in water? How did Frank Zappa die? Name the Baltic countries. Which seabird was declared extinct in the 1840s? Who is Noam Chomsky? List names of Russian composers. Edison is the inventor of what? How far is the moon from the sun? What is the distance from New York to Boston? How many planets are there? What is the exchange rate of the Euro to the Dollar? What does SPQR stand for? What is the nickname of Totti? What does the Scottish word “bonnie” mean? Who wrote the song “Paranoid Android”?