Summarization and Generation

1

Summarization and Summarization and GenerationGeneration

CS 4705

2

What is Summarization?What is Summarization?What is Summarization?What is Summarization? Data as input (database, software trace, expert

system), text summary as output

Text as input (one or more articles), paragraph summary as output

Multimedia in input or output

Summaries must convey maximal information in minimal space

3

Summarization is not the same Summarization is not the same as Language Generationas Language Generation Karl Malone scored 39 points Friday

night as the Utah Jazz defeated the Boston Celtics 118-94.

Karl Malone tied a season high with 39 points Friday night….

… the Utah Jazz handed the Boston Celtics their sixth straight home defeat 118-94.

Streak, Jacques Robin, 1993

4

Summarization TasksSummarization Tasks

Linguistic summarization: How to pack in as much information as possible in as short an amount of space as possible?

Streak: Jacques Robin MAGIC: James Shaw PLanDoc: Karen Kukich, James Shaw, Rebecca

Passonneau, Hongyan Jing, Vasilis Hatzivassiloglou

Conceptual summarization: What information should be included in the summary?

5

Input Data -- STREAKInput Data -- STREAKInput Data -- STREAKInput Data -- STREAK

score (Jazz, 118)score (Celtics, 94)

The Utah Jazz beat theCeltics 118 - 94.

points (Malone, 39) Karl Malone scored 39points

location(game,Boston)

It was a home gamefor the Celtics

#home-defeats(Celtics, 6)

It was the 6th straighthome defeat

6

Revision ruleRevision ruleRevision ruleRevision rule

beatbeat

JazzJazz CelticsCeltics

handhand

JazzJazz defeatdefeat CelticsCeltics

7

SUMMONS QUERY OUTPUT

Summary:

Wednesday, April 19, 1995, CNN reported that anexplosion shook a government building inOklahoma City. Reuters announced that at least 18people were killed. At 1 PM, Reuters announcedthat three males of Middle Eastern origin werepossibly responsible for the blast. Two days later,Timothy McVeigh, 27, was arrested as a suspect,U.S. attorney general Janet Reno said. As of May29, 1995, the number of victims was 166.

Image(s):

1 (okfed1.gif) (WebSeek)

Article(s):(1) Blast hits Oklahoma Citybuilding

(2) Suspects' truck said rented from Dallas

(3) At least 18 killed in bombblast - CNN

(4) DETROIT (Reuter) - A federal judgeMonday ordered James

(5) WASHINGTON (Reuter) - Asuspect in the Oklahoma Citybombing

Summons, Dragomir Radev, 1995

8

BriefingsBriefingsBriefingsBriefings

Transitional Automatically summarize series of articles Input = templates from information extraction Merge information of interest to the user from

multiple sources Show how perception changes over time Highlight agreement and contradictions

Conceptual summarization: planning operators Refinement (number of victims) Addition (Later template contains perpetrator)

9

How is summarization done?How is summarization done?How is summarization done?How is summarization done?

4 input articles parsed by information extraction system

4 sets of templates produced as output Content planner uses planning operators

to identify similarities and trends Refinement (Later template reports new # victims)

New template constructed and passed to sentence generator

10

Sample TemplateSample TemplateSample TemplateSample Template

Message ID TST-COL-0001

Secsource: source ReutersSecsource: date 26 Feb 93

Early afternoonIncident: date 26 Feb 93Incident: location World Trade CenterIncident:Type BombingHum Tgt: number At least 5

11

Document SummarizationDocument Summarization

Input: one or more text documents

Output: paragraph length summary

Sentence extraction is the standard method Using features such as key words, sentence position in document, cue phrases Identify sentences within documents that are salient Extract and string sentences together Luhn – 1950s Hovy and Lin 1990s Schiffman 2000

Machine learning for extraction Corpus of document/summary pairs Learn the features that best determine important sentences Kupiec 1995: Summarization of scientific articles

12

Summarization ProcessSummarization Process

Shallow analysis instead of information extraction

Extraction of phrases rather than sentences

Generation from surface representations in place of semantics

13

Problems with Sentence Problems with Sentence ExtractionExtraction Extraneous phrases

“The five were apprehended along Interstate 95, heading south in vehicles containing an array of gear including … ... authorities said.”

Dangling noun phrases and pronouns “The five”

MisleadingWhy would the media use this specific word

(fundamentalists), so often with relation to Muslims? *Most of them are radical Baptists, Lutheran and Presbyterian groups.

14

Cut and PasteCut and Paste in Professional in Professional SummarizationSummarization

Humans also reuse the input text to produce summaries

But they “cut and paste” the input rather than simply extract our automatic corpus analysis

300 summaries, 1,642 sentences 81% sentences were constructed by cutting

and pasting linguistic studies

15

Major Cut and Paste Major Cut and Paste OperationsOperations (1) Sentence reduction

~~~~~~~~~~~~

16


~~~~~~~~~~~~

17


(2) Sentence Combination

~~~~~~~~~~~~

~~~~~~~~~~~~~~ ~~~~~~

18

Major Cut and Paste Major Cut and Paste OperationsOperations (3) Syntactic Transformation

(4) Lexical paraphrasing

~~~~~

~~~~~~~~~~~

~~~~~

~~~

19

Summarization at ColumbiaSummarization at Columbia

News: Newsblaster, GALE Email Meetings Journal articles Open-ended question-answering

What is a Loya Jurga? Who is Mohammed Naeem Noor Khan? What do people think of welfare reform?

20

Summarization at ColumbiaSummarization at Columbia

News Single Document Multi-document

Email Meetings Journal articles Open-ended question-answering

What is a Loya Jurga? Who is Al Sadr? What do people think of welfare reform?

21

Cut and Paste Based Single Document Cut and Paste Based Single Document Summarization -- System ArchitectureSummarization -- System Architecture

Extraction

Sentence reduction

Generation

Sentence combination

Input: single document

Extracted sentences

Output: summary

Corpus

Decomposition

Lexicon

Parser

Co-reference

22

(1) Decomposition of Human-(1) Decomposition of Human-written Summary Sentenceswritten Summary Sentences

Input: a human-written summary sentence the original document

Decomposition analyzes how the summary sentence was constructed

The need for decomposition provide training and testing data for

studying cut and paste operations

23

Sample Decomposition OutputSample Decomposition OutputSummary sentence: Arthur B. Sackler, vice

president for law and public policy of Time Warner Cable Inc. and a member of the direct marketing association told the Communications Subcommittee of the Senate Commerce Committee that legislation to protect children’s privacy on-line could destroy the spondtaneous nature that

makes the Internet unique.

Document sentences:S1: A proposed new law that would require web publishers to obtain parental consent before collecting personal information from children could destroy the spontaneous nature that makes the internet unique, a member of the Direct Marketing Association told a Senate panel Thursday.S2: Arthur B. Sackler, vice president for law and public policy of Time Warner Cable Inc., said the association supported efforts to protect children on-line, but he…S3: “For example, a child’s e-mail address is necessary …,” Sackler said in testimony to the Communications subcommittee of the SenateCommerce Committee.S5: The subcommittee is considering the Children’s Online Privacy Act, which was drafted…

24

Decomposition of human-written Decomposition of human-written summariessummaries

Summary sentence: Arthur B. Sackler, vice




A Sample Decomposition OutputA Sample Decomposition Output

25







26







27

Algorithm for DecompositionAlgorithm for Decomposition A Hidden Markov Model based solution Evaluations:

Human judgements 50 summaries, 305 sentences 93.8% of the sentences were decomposed

correctly Summary sentence alignment Tested in a legal domain

Details in (Jing&McKeown-SIGIR99)

28

(2) Sentence Reduction(2) Sentence Reduction

An example:Original Sentence: When it arrives sometime next year in

new TV sets, the V-chip will give parents a new and potentially revolutionary device to block out programs they don’t want their children to see.

Reduction Program: The V-chip will give parents a new and potentially revolutionary device to block out programs they don’t want their children to see.

Professional: The V-chip will give parents a device to block out programs they don’t want their children to see.

29

Algorithm for Sentence Algorithm for Sentence ReductionReduction Preprocess: syntactic parsing Step 1: Use linguistic knowledge to decide

what phrases MUST NOT be removed Step 2: Determine what phrases are

most important in the local context Step 3: Compute the probabilities of

humans removing a certain type of phrase

Step 4: Make the final decision

30

Step 1: Use linguistic knowledge Step 1: Use linguistic knowledge to decide what MUST NOT be to decide what MUST NOT be removedremoved

Syntactic knowledge from a large-scale, reusable lexicon convince: meaning 1: NP-PP :PVAL (“of”) (E.g., “He convinced me of his innocence”)

NP-TO-INF-OC (E.g., “He convinced me to go to the party”)

meaning 2: ... Required syntactic arguments are not

removed

31

Saudi Arabia on Tuesday decided to sign… The official Saudi Press Agency reported

that King Fahd made the decision during a cabinet meeting in Riyadh, the Saudi capital.

The meeting was called in response to … the Saudi foreign minister, that the Kingdom…

An account of the Cabinet discussions and decisions at the meeting…

The agency...

Step 2: Determining context Step 2: Determining context importance based on lexical importance based on lexical linkslinks

32





The agency...

Step 2: Determining context Step 2: Determining context importance based on lexical importance based on lexical links links

33





The agency...

Step 2: Determining context Step 2: Determining context importance based on lexical importance based on lexical links links

34

Step 3: Compute probabilities Step 3: Compute probabilities of humans removing a phraseof humans removing a phrase

verb(will give)

vsubc(when)

subj(V-chip)

iobj(parents)

obj(device)

ndet(a)

adjp(and)

lconj(new)

rconj(revolutionary)

Prob(“when_clause is removed”| “v=give”)

Prob (“to_infinitive modifier is removed” | “n=device”)

35

Step 4: Make the final Step 4: Make the final decisiondecision

verb(will give)

vsubc(when)

subj(V-chip)

iobj(parents)

ndet(a)

adjp(and)

lconj(new)

rconj(revolutionary)

L Cn Pr

L Cn Pr

L Cn Pr L Cn Pr L Cn Pr

L Cn Pr

L Cn Pr L Cn Pr

obj(device)

L Cn Pr

L -- linguisticCn -- contextPr -- probabilities

36

Evaluation of ReductionEvaluation of Reduction

Success rate: 81.3% 500 sentences reduced by humans Baseline: 43.2% (remove all the clauses,

prepositional phrases, to-infinitives,…)

Reduction rate: 32.7% Professionals: 41.8%

Details in (Jing-ANLP00)

37

Multi-Document Summarization Multi-Document Summarization Research FocusResearch Focus Monitor variety of online information sources

News, multilingual Email

Gather information on events across source and time

Same day, multiple sources Across time

Summarize Highlighting similarities, new information, different

perspectives, user specified interests in real-time

38

ApproachApproach

Use a hybrid of statistical and linguistic knowledge

Statistical analysis of multiple documents Identify important new, contradictory information

Information fusion and rule-driven content selection

Generation of summary sentences By re-using phrases Automatic editing/rewriting summary

39

NewsblasterNewsblaster

http://newsblaster.cs.columbia.edu/

Clustering articles into events Categorization by broad topic Multi-document summarization Generation of summary sentences

Fusion Editing of references

40

Newsblaster Newsblaster ArchitectureArchitecture

Crawl News SitesCrawl News Sites

Form ClustersForm Clusters

CategorizeCategorize

TitleClusters

TitleClusters

SummaryRouter

SummaryRouter

SelectImagesSelectImages

EventSummary

EventSummary

BiographySummaryBiographySummary

Multi-EventMulti-Event

Convert Output to HTMLConvert Output to HTML

42

Fusion Fusion

43

Sentence Fusion ComputationSentence Fusion Computation

Common information identification Alignment of constituents in parsed theme sentences: only some

subtrees match Bottom-up local multi-sequence alignment Similarity depends on

Word/paraphrase similarity Tree structure similarity

Fusion lattice computation Choose a basis sentence Add subtrees from fusion not present in basis Add alternative verbalizations Remove subtrees from basis not present in fusion

Lattice linearization Generate all possible sentences from the fusion lattice Score sentences using statistical language model

46

Tracking Across DaysTracking Across Days

Users want to follow a story across time and watch it unfold

Network model for connecting clusters across days Separately cluster events from today’s news Connect new clusters with yesterday’s news Allows for forking and merging of stories

Interface for viewing connections Summaries that update a user on what’s new

Statistical metrics to identify differences between article pairs Uses learned model of features Identifies differences at clause and paragraph levels

51

Different PerspectivesDifferent Perspectives

Hierarchical clustering Each event cluster is divided into clusters by

country

Different perspectives can be viewed side by side

Experimenting with update summarizer to identify key differences between sets of stories

55

Multilingual SummarizationMultilingual Summarization

Given a set of documents on the same event

Some documents are in English

Some documents are translated from other languages

56

Issues for Multilingual Issues for Multilingual SummarizationSummarization Problem: Translated text is errorful

Exploit information available during summarization

Similar documents in cluster

Replace translated sentences with similar English

Edit translated text Replace named entities with extractions from similar English

57

Multilingual RedundancyMultilingual RedundancyBAGDAD. - A total of 21 prisoners has been died and a hundred more hurt by firings from mortar in the jail of Abu Gharib (to 20 kilometers to the west of Bagdad), according to has informed general into the U.S.A. Marco Kimmitt.

Bagdad in the Iraqi capital Aufstaendi attacked Bagdad on Tuesday a prison with mortars and killed after USA gifts 22 prisoners. Further 92 passengers of the Abu Ghraib prison were hurt, communicated a spokeswoman of the American armed forces.

The Iraqi being stationed US military shot on the 20th, the same day to the allied forces detention facility which is in アブグレイブ of the Baghdad west approximately 20 kilometers, mortar 12 shot and you were packed, 22 Iraqi human prisoners died, it announced that nearly 100 people were injured.

BAGHDAD, Iraq – Insurgents fired 12 mortars into Baghdad's Abu Ghraib prison Tuesday, killing 22 detainees and injuring 92, U.S. military officials said.

Germ

anG

erman

Spanish

Spanish

JapaneseJapanese

English

English

58

Multilingual RedundancyMultilingual RedundancyBAGDAD. - A total of 21 prisoners has been died and a hundred more hurt by firings from mortar in the jail of Abu Gharib (to 20 kilometers to the west of Bagdad), according to has informed general into the U.S.A. Marco Kimmitt.

Bagdad in the Iraqi capital Aufstaendi attacked Bagdad on Tuesday a prison with mortars and killed after USA gifts 22 prisoners. Further 92 passengers of the Abu Ghraib prison were hurt, communicated a spokeswoman of the American armed forces.

The Iraqi being stationed US military shot on the 20th, the same day to the allied forces detention facility which is in アブグレイブ of the Baghdad west approximately 20 kilometers, mortar 12 shot and you were packed, 22 Iraqi human prisoners died, it announced that nearly 100 people were injured.

BAGHDAD, Iraq – Insurgents fired 12 mortars into Baghdad's Abu Ghraib prison Tuesday, killing 22 detainees and injuring 92, U.S. military officials said.

Germ

anG

erman

Spanish

Spanish

JapaneseJapanese

English

English

59

Multilingual Similarity-based Multilingual Similarity-based SummarizationSummarization

ستجوا ب

القالفDeutsch

erText

日本語テキスト

EnglishTextEnglish

TextEnglishText

Related English Documents

Documents on the same Event

MachineTranslation

ArabicText

German

Text

Japanese

Text

DetectSimilar

Sentences

Replace / rewrite MT sentences

EnglishSumma

ry

Select summary sentences

Simplify English

Sentences

60

Similarity ComputationSimilarity Computation

Simfinder computes similarity between sentences based on multiple features: Proper Nouns Verb, noun, adjective WordNet (synonyms) word stem overlap

New: Noun phrase and noun phrase variant feature

(FASTR)

61

Sentence 1Iraqi President Saddam Hussein that the government of Iraq

over 24 years in a "black" near the port of the northern Iraq after nearly eight months of pursuit was considered the largest in history .

Similarity 0.27: Ousted Iraqi President Saddam Hussein is in custody following his dramatic capture by US forces in Iraq.

Similarity 0.07: Saddam Hussein, the former president of Iraq, has been captured and is being held by US forces in the country.

Similarity 0.04: Coalition authorities have said that the former Iraqi president could be tried at a war crimes tribunal, with Iraqi judges presiding and international legal experts acting as advisers.

62

Sentence SimplificationSentence Simplification

Machine translated sentences long and ungrammatical

Use sentence simplification on English sentences to reduce to approximately “one fact” per sentence

Use Arabic sentences to find most similar simple sentences

Present multiple high similarity sentences

63

Simplification ExamplesSimplification Examples 'Operation Red Dawn', which led to the capture of

Saddam Hussein, followed crucial information from a member of a family close to the former Iraqi leader. ' Operation Red Dawn' followed crucial information from a

member of a family close to the former Iraqi leader. Operation Red Dawn led to the capture of Saddam Hussein.

Saddam Hussein had been the object of intensive searches by US-led forces in Iraq but previous attempts to locate him had proved unsuccessful. Saddam Hussein had been the object of intensive searches

by US-led forces in Iraq. But previous attempts to locate him had proved unsuccessful.

64

Results on alquds.co.uk.195Results on alquds.co.uk.195

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.90

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

F-measure vs. Threshold

Full Threshold .10Full Threshold .65Simplified Threshold .10Simplified Threshold .65FASTR Simplified .10FASTR Simplified .65

Threshold

F-m

ea

sure

68

EvaluationEvaluation

DUC (Document Understanding Conference): run by NIST

Held annually Manual creation of topics (sets of documents) 2-7 human written summaries per topic How well does a system generated summary

cover the information in a human summary?

69

User Study: ObjectivesUser Study: Objectives

Does multi-document summarization help?

Do summaries help the user find information needed to perform a report writing task?

Do users use information from summaries in gathering their facts?

Do summaries increase user satisfaction with the online news system?

Do users create better quality reports with summaries? How do full multi-document summaries compare with

minimal 1-sentence summaries such as Google News?

70

User Study: DesignUser Study: Design

Four parallel news systems Source documents only; no summaries Minimal single sentence summaries (Google

News) Newsblaster summaries Human summaries

All groups write reports given four scenarios A task similar to analysts Can only use Newsblaster for research Time-restricted

71

User Study: ExecutionUser Study: Execution

4 scenarios 4 event clusters each 2 directly relevant, 2 peripherally relevant Average 10 documents/cluster

45 participants Balance between liberal arts, engineering 138 reports

Exit survey Multiple-choice and open-ended questions

Usage tracking Each click logged, on or off-site

72

““Geneva” PromptGeneva” Prompt

The conflict between Israel and the Palestinians has been difficult for government negotiators to settle. Most recently, implementation of the “road map for peace”, a diplomatic effort sponsored by ……

Who participated in the negotiations that produced the Geneva Accord?

Apart from direct participants, who supported the Geneva Accord preparations and how?

What has the response been to the Geneva Accord by the Palestinians?

73

Measuring EffectivenessMeasuring Effectiveness

Score report content and compare across summary conditions

Compare user satisfaction per summary condition

Comparing where subjects took report content from

75

User SatisfactionUser Satisfaction

More effective than a web search with Newsblaster

Not true with documents only or single-sentence summaries

Easier to complete the task with summaries than with documents only

Enough time with summaries than documents only

Summaries helped most 5% single sentence summaries 24% Newsblaster summaries 43% human summaries

76

User Study: ConclusionsUser Study: Conclusions

Summaries measurably improve a news browswer’s effectiveness for research

Users are more satisfied with Newsblaster summaries are better than single-sentence summaries like those of Google News

Users want search Not included in evaluation

77

Email SummarizationEmail Summarization

Cross between speech and text Elements of dialog Informal language More context explicitly repeated than speech

Wide variety of types of email Conversation to decision-making

Different reasons for summarization Browsing large quantities of email – a mailbox Catch-up: join a discussion late and participate – a

thread

78

Email Summarization: ApproachEmail Summarization: Approach

Collected and annotated multiple corpora of email Hand-written summary, categorization threads&messages Identified 3 categories of email to address:

Event planning, Scheduling, Information gathering

Developed tools: Automatic categorization of email Preliminary summarizers

Statistical extraction using email specific features Components of category specific summarization

79

Email Summarization by Email Summarization by Sentence ExtractionSentence Extraction

Use features to identify key sentences Non-email specific: e.g., similarity to centroid Email specific: e.g., following quoted material

Rule-based supervised machine learning Training on human-generated summaries

Add “wrappers” around sentences to show who said what

80

Data for Sentence ExtractionData for Sentence Extraction Columbia ACM chapter executive board mailing list Approximately 10 regular participants ~300 Threads, ~1000 Messages Threads include: scheduling and planning of meetings

and events, question and answer, general discussion and chat.

Annotated by human annotators: Hand-written summary Categorization of threads and messages Highlighting important information (such as question-answer

pairs)

81

Email Summarization by Email Summarization by Sentence ExtractionSentence Extraction

Creation of Training Data Start with human-generated summaries Use SimFinder (a trained sentence similarity measure –

Hatzivassiloglou et al 2001) to label sentences in threads as important

Learning of Sentence Extraction Rules Use Ripper (a rule learning algorithm – Cohen 1996) to learn rules

for sentence classification Use basic and email-specific features in machine learning

Creating summaries Run learned rules on unseen data Add “wrappers” around sentences to show who said what

Results Basic: .55 precision; .40 F-measure Email-specific: .61 precision; .50 F-measure

82

Sample Automatically Generated Sample Automatically Generated Summary (ACM0100)Summary (ACM0100)

Regarding "meeting tonight...", on Oct 30, 2000, David Michael Kalin wrote: Can I reschedule my C session for Wednesday night, 11/8, at 8:00?

Responding to this on Oct 30, 2000, James J Peach wrote: Are you sure you want to do it then?

Responding to this on Oct 30, 2000, Christy Lauridsen wrote: David , a reminder that your scheduled to do an MSOffice session on Nov. 14, at 7pm in 252Mudd.

83

Information Gathering Email:Information Gathering Email:The ProblemThe ProblemSummary from our rule-based sentence extractor:

Regarding "acm home/bjarney", on Apr 9, 2001, Mabel Dannon wrote:Two things: Can someone be responsible for the press releases for Stroustrup?

Responding to this on Apr 10, 2001, Tina Ferrari wrote:I think Peter, who is probably a better writer than most of us, is writing up something for dang and Dave to send out to various ACM chapters. Peter, we can just use that as our "press release", right?

In another subthread, on Apr 12, 2001, Keith Durban wrote:Are you sending out upcoming events for this week?

84

Detection of QuestionsDetection of Questions

Questions in interrogative form: inverted subject-verb order

Supervised rule induction approach, training Switchboard, test ACM corpus

Results:

Recall low because:

Questions in ACM corpus start with a declarative clause So, if you're available, do you want to come? if you don't mind, could you post this to the class bboard?

Results without declarative-initial questions:

Recall 0.56

Precision 0.96

F-measure 0.70

Recall 0.72

Precision 0.96

F-measure 0.82

85

Detection of AnswersDetection of AnswersSupervised Machine Learning Approach Use human annotated data to generate gold standard

training data Annotators were asked to highlight and associate question-answer

pairs in the ACM corpus.

Learn a classifier that predicts if a subsequent segment to a question segment answers it Represent each question and candidate answer segment

by a feature vectorLabeller 1

Labeller 2

Union

Precision 0.690 0.680 0.728

Recall 0.652 0.612 0.732

F1-Score 0.671 0.644 0.730

86

Integrating QA detection with Integrating QA detection with summarizationsummarization Use QA labels as features in sentence

extraction (F=.545) Add automatically detected answers to

questions in extractive summaries (F=.566)

Start with QA pair sentences and augmented with extracted sentences (F=.573)

87

Integrated in Microsoft OutlookIntegrated in Microsoft Outlook

88

Meeting SummarizationMeeting Summarization(joint with Berkeley, SRI, Washington)(joint with Berkeley, SRI, Washington)

Goal: automatic summarization of meetings by generating “minutes” highlighting the debate that affected each decision.

Work to date: Identification of agreement/disagreement Machine learning approach: lexical, structure,

acoustic features Use of context: who agreed with who so far?

Adressee identification Bayesian modeling of context

89

ConclusionsConclusions

Non-extractive summarization is practical today

User studies show summarization improves access to needed information

Advances and ongoing research in tracking events, multilingual summarization, perspective identification

Moves to new media (email, meetings) raise new challenges with dialog, informal language

Summarization and Generation

Documents

Transcript of Summarization and Generation