Post on 25-Feb-2016
description
TEXT PROCESSING 1
Anaphora resolutionIntroduction to Anaphora Resolution
Outline
A reminder: anaphora resolution, factors affecting the interpretation of anaphoric expressions
A brief history of anaphora resolution First algorithms: Charniak, Winograd, Wilks Pronouns: Hobbs Salience: S-List, LRC
Early ML work Definite descriptions: Vieira & Poesio The MUC initiative – also: coreference evaluation
methods Soon et al
Anaphora resolution: a specification of the problem
Anaphora resolution:coreference chains
Reminder: Factors that affect the interpretation of anaphoric expressions
Factors: Morphological features (agreement) Syntactic information Salience Lexical and commonsense knowledge
Distinction often made between CONSTRAINTS and PREFERENCES
Agreement
GENDER strong CONSTRAINT for pronouns (in other languages: for other anaphors as well) [Jane] blamed [Bill] because HE spilt the
coffee (Ehrlich, Garnham e.a, Arnold e.a) NUMBER also strong constraint
[[Union] representatives] told [the CEO] that THEY couldn’t be reached
Some complexities
Gender: [India] withdrew HER ambassador from the
Commonwealth “…to get a customer’s 1100 parcel-a-week load
to its doorstep” [actual error from LRC algorithm]
Number: The Union said that THEY would withdraw from
negotations until further notice.
Syntactic information
BINDING constraints [John] likes HIM
EMBEDDING constraints [[his] friend]
PARALLELISM preferences Around 60% of pronouns occur in subject
position; around 70% of those refer to antecedents in subject position
[John] gave [Bill] a book, and [Fred] gave HIM a pencil
Effect of syntax on SALIENCE (Next)
Salience
In every discourse, certain entities are more PROMINENT
Factors that affect prominence
Distance Order of mention in the sentence
Entities mentioned earlier in the sentence more prominent Type of NP (proper names > other types of NPs) Number of mentions Syntactic position (subj > other GF, matrix >
embedded) Semantic role (‘implicit causality’ theories) Discourse structure
Focusing theories
Hypothesis: One or more entities in the discourse are the FOCUS OF (LINGUISTIC) ATTENTION just like some entities in the visual space are the focus of VISUAL attention Grosz 1977, Reichman 1985: ‘focus
spaces’ Sidner 1979, Sanford & Garrod 1981:
‘focused entities’ Grosz et al 1981, 1983, 1995: Centering
Lexical and commonsense knowledge
[The city council] refused [the women] a permit because they feared violence.[The city council] refused [the women] a permit because they advocated violence.
Winograd (1974), Sidner (1979)
BRISBANE – a terrific right rip from [Hector Thompson] dropped [Ross Eadie] at Sandgate on Friday night and won him the Australian welterweight boxing title. (Hirst, 1981)
Problems to be resolved by an AR system: mention identification
Effect: recall Typical problems:
Nested NPs (possessives) [a city] 's [computer system] [[a city]’s computer system]
Appositions: [Madras], [India] [Madras, [India]]
Attachments
Problems for AR:Complex attachments
[The quality that’s coming out of [software from [India]] The quality that’s coming out of software
from India is now exceeding the quality of software that’s coming out from the United States
scanning through millions of lines of computer code
ACE/bnews/devel/ABC19981001.1830.1257
Problems for AR: agreement extraction
The committee are meeting / is meeting
The Union sent a representative. They ….
The doctor came to visit my father. SHE told him …
Problems to be solved: anaphoricity determination
Expletives: IT’s not easy to find a solution Is THERE any reason to be optimistic at
all? Non-anaphoric definites
Outline
A reminder: anaphora resolution, factors affecting the interpretation of anaphoric expressions
A brief history of anaphora resolution First algorithms: Charniak, Winograd, Wilks Pronouns: Hobbs Salience: S-List, LRC
Early ML work Definite descriptions: Vieira & Poesio The MUC initiative, coreference evaluation methods Soon et al 2001
A brief history of early work on anaphora resolution
Algorithms heavily based on (hand-coded) commonsense knowledge: Charniak 72, Winograd 74, Hobbs 1974, Wilks 1975
Syntax-based algorithms: Hobbs 1974 Salience-based algorithms: Brennan et
al, Strube & Hahn, Tetreault
First approaches to anaphora resolution: Charniak, Winograd, Wilks
Heavy emphasis on inference Hand-coded No formal evaluation
1970s Example: SHRDLUhttp://hci.stanford.edu/~winograd/shrdlu/
Person: PICK UP A BIG RED BLOCK. Computer: OK. (does it) Person: GRASP THE PYRAMID. Computer: I DON'T UNDERSTAND WHICH PYRAMID YOU MEAN. Person: FIND A BLOCK WHICH IS TALLER THAN THE ONE YOU ARE HOLDING AND PUT IT INTO THE BOX. Computer: BY "IT", I ASSUME YOU MEAN THE BLOCK WHICH IS
TALLER THAN THE ONE I AM HOLDING. Computer: OK. (does it) Person: WHAT DOES THE BOX CONTAIN? Computer: THE BLUE PYRAMID AND THE BLUE BLOCK. Person: WHAT IS THE PYRAMID SUPPORTED BY? Computer: THE BOX. Person: HOW MANY BLOCKS ARE NOT IN THE BOX? Computer: FOUR OF THEM. Person: IS AT LEAST ONE OF THEM NARROWER THAN THE ONE WHICH I TOLD YOU TO PICK UP? Computer: YES, THE RED CUBE.
• This is one of the only dialogs it knows.• SHRDLU is too stupid to make mistakes.• Beautiful “Demo-ware”
Terry Winograd. 1971.MIT Ph.D. Thesis. Terry Winograd
Anaphora in SHRDLU
First example of HISTORY LIST algorithm Uses a combination of agreement features and
semantic constraints Check all possibilities and assign a global score
rather than simply find the first match Score incorporates syn component: entities in subj
position higher score than entities in object position, in turn ranked more highly than entities in adjunct position
Performance made more impressive by including solutions to a number of complex cases, such as reference to events (Why did you do it?) – often ad hoc
Hobbs’ `Naïve Algorithm’ (Hobbs, 1974)
The reference algorithm for PRONOUN resolution (until Soon et al it was the standard baseline) Interesting since Hobbs himself in the
1974 paper suggests that this algorithm is very limited (and proposes one based on semantics)
The first anaphora resolution algorithm to have an (informal) evaluation
Purely syntax based
Hobbs’ `Naïve Algorithm’ (Hobbs, 1974)
Works off ‘surface parse tree’ Starting from the position of the pronoun in
the surface tree, first go up the tree looking for an antecedent in
the current sentence (left-to-right, breadth-first);
then go to the previous sentence, again traversing left-to-right, breadth-first.
And keep going back
Hobbs’ algorithm: Intrasentential anaphora
Steps 2 and 3 deal with intrasentential anaphora and incorporate basic syntactic constraints:
Also: John’s portrait of him
S
NPJohn V
likesNPhim
Xp
Hobbs’ Algorithm: intersentential anaphora
S
NPJohn V
likesNPhim
S
NPBill V
isNPa good friend
X
candidate
Evaluation The first anaphora resolution algorithm to be evaluated in a
systematic manner, and still often used as baseline (hard to beat!) Hobbs, 1974:
300 pronouns from texts in three different styles (a fiction book, a non-fiction book, a magazine)
Results: 88.3% correct without selectional constraints, 91.7% with SR 132 ambiguous pronouns; 98 correctly resolved.
Tetreault 2001 (no selectional restrictions; all pronouns) 1298 out of 1500 pronouns from 195 NYT articles (76.8% correct) 74.2% correct intra, 82% inter
Main limitations Reference to propositions excluded Plurals Reference to events
Salience-based algorithms
Common hypotheses: Entities in discourse model are RANKED by
salience Salience gets continuously updated Most highly ranked entities are preferred
antecedents Variants:
DISCRETE theories (Sidner, Brennan et al, Strube & Hahn): 1-2 entities singled out
CONTINUOUS theories (Alshawi, Lappin & Leass, Strube 1998, LRC): only ranking
Salience-based algorithms
Sidner 1979: Most extensive theory of the influence of salience on
several types of anaphors Two FOCI: discourse focus, agent focus never properly evaluated
Brennan et al 1987 (see Walker 1989) Ranking based on grammatical function One focus (CB)
Strube & Hahn 1999 Ranking based on information status (NP type)
S-List (Strube 1998): drop CB LRC (Tetreault): incremental
LRC
An update on Strube’s S-LIST algorithm (= Centering without centers)
Initial version augmented with various syntactic and discourse constraints
LRC Algorithm (LRC)
Maintain a stack of entities ranked by grammatical function and sentence order (Subj > DO > IO)
Each sentence is represented by a Cf-list: list of salient entities ordered by grammatical function
While processing utterance’s entities (left to right) do: Push entity onto temporary list (Cf-list-new), if pronoun, attempt
to resolve first: Search through Cf-list-new (l-to-r) taking the first candidate
that meets gender, agreement constraints, etc. If none found, search past utterance’s Cf-lists starting from
previous utterance to beginning of discourse
Results
Algorithm PTB-News (1694)
PTB-Fic (511)
LRC 74.9% 72.1%
S-List 71.7% 66.1%
BFP 59.4% 46.4%
Comparison with ML techniques of the time
Algorithm All 3rd
LRC 76.7%
Ge et al. (1998) 87.5% (*)
Morton (2000) 79.1%
Carter’s Algorithm (1985)
The most systematic attempt to produce a system using both salience (Sidner’s theory) & commonsense knowledge (Wilks’s preference semantics)
Small-scale evaluation (around 100 hand-constructed examples)
Many ideas found their way in the SRI’s Core Language Engine (Alshawi, 1992)
Outline
A reminder: anaphora resolution, factors affecting the interpretation of anaphoric expressions
A brief history of anaphora resolution First algorithms: Charniak, Winograd, Wilks Pronouns: Hobbs Salience: S-List, LRC
Modern work in anaphora resolution Early ML work The MUC initiative – also: coreference evaluation
methods Soon et al
MODERN WORK IN ANAPHORA RESOLUTION
Availability of the first anaphorically annotated corpora circa 1993 (MUC6) made statistical methods possible
STATISTICAL APPROACHES TO ANAPHORA RESOLUTION
UNSUPERVISED approaches Eg Cardie & Wagstaff 1999, Ng 2008
SUPERVISED approaches Early (NP type specific) Soon et al: general classifier + modern
architecture
ANAPHORA RESOLUTION AS A CLASSIFICATION PROBLEM
1. Classify NP1 and NP2 as coreferential or not
2. Build a complete coreferential chain
SOME KEY DECISIONS
ENCODING I.e., what positive and negative instances to
generate from the annotated corpus Eg treat all elements of the coref chain as
positive instances, everything else as negative: DECODING
How to use the classifier to choose an antecedent
Some options: ‘sequential’ (stop at the first positive), ‘parallel’ (compare several options)
Outline
A reminder: anaphora resolution, factors affecting the interpretation of anaphoric expressions
A brief history of anaphora resolution First algorithms: Charniak, Winograd, Wilks Pronouns: Hobbs Salience: S-List, LRC
Modern work in anaphora resolution Early ML work The MUC initiative – also: coreference evaluation
methods Soon et al
Early machine-learning approaches
Main distinguishing feature: concentrate on a single NP type
Both hand-coded and ML: Aone & Bennett (pronouns) Vieira & Poesio (definite descriptions)
Ge and Charniak (pronouns)
Definite descriptions:Vieira & Poesio
A first attempt at going beyond pronouns while still doing a large-scale evaluation
Definite descriptions chosen because they require lexical and commonsense knowledge
Developing both a hand-coded and a ML decision tree (as in Aone and Bennett)
Vieira & Poesio 1996, Vieira 1998, Vieira & Poesio 2000
Preliminary corpus study (Poesio and Vieira, 1998)
Annotators asked to classify about 1,000 definite descriptions from the ACL/DCI corpus (Wall Street Journal texts) into three classes:
DIRECT ANAPHORA: a house … the house
DISCOURSE-NEW: the belief that ginseng tastes like spinach is more widespread than one would expect
BRIDGING DESCRIPTIONS:the flat … the living room; the car … the vehicle
Poesio and Vieira, 1998
Results:
More than half of the def descriptions are first-mention
Subjects didn’t always agree on the classification of an antecedent (bridging descriptions: ~8%)
The Vieira / Poesio system for robust definite description resolution
Follows a SHALLOW PROCESSING approach (Carter, 1987; Mitkov, 1998): it only uses
Structural information (extracted from Penn Treebank)
Existing lexical sources (WordNet)
(Very little) hand-coded information
Methods for resolving direct anaphors
DIRECT ANAPHORA:
the red car, the car, the blue car: premodification heuristics
segmentation: approximated with ‘loose’ windows
Methods for resolving discourse-new definite descriptions
DISCOURSE-NEW DEFINITES
the first man on the Moon, the fact that Ginseng tastes of spinach: a list of the most common functional predicates (fact, result, belief) and modifiers (first, last, only… )
heuristics based on structural information (e.g., establishing relative clauses)
The (hand-coded) decision tree
1. Apply ‘safe’ discourse-new recognition heuristics2. Attempt to resolve as same-head anaphora3. Attempt to classify as discourse new4. Attempt to resolve as bridging description.
Search backward 1 sentence at a time and apply heuristics in the following order:1. Named entity recognition heuristics – R=.66, P=.952. Heuristics for identifying compound nouns acting as
anchors – R=.363. Access WordNet – R, P about .28
The decision tree obtained via ML
Same features as for the hand-coded decision tree
Using ID3 classifier (non probabilistic decision tree)
Training instances: Positive: closest annotated antecedent Negative: all mentions in previous four
sentences Decoding: consider all possible
antecedents in previous four sentences
Automatically learned decision tree
Overall Results
Evaluated on a test corpus of 464 definite descriptions
Overall results:
R P FVersion 1 53% 76% 62%Version 2 57% 70% 62%D-N def 77% 77% 77%ID3 75% 75% 75%
Overall Results
Results for each type of definite description:
R P FDirect anaphora 62% 83% 71%
Disc new 69% 72% 70%Bridging 29% 38% 32.9%
Outline
A reminder: anaphora resolution, factors affecting the interpretation of anaphoric expressions
A brief history of anaphora resolution First algorithms: Charniak, Winograd, Wilks Pronouns: Hobbs Salience: S-List, LRC
Early ML work Definite descriptions: Vieira & Poesio The MUC initiative and coreference evaluation
methods Soon et al
MUC
First big initiative in Information Extraction
Produced first sizeable annotated data for coreference
Developed first methods for evaluating systems
MUC terminology:
MENTION: any markable COREFERENCE CHAIN: a set of
mentions referring to an entity KEY: the (annotated) solution (a
partition of the mentions into coreference chains)
RESPONSE: the coreference chains produced by a system
Evaluation of coreference resolution systems
Lots of different measures proposed ACCURACY:
Consider a mention correctly resolved if Correctly classified as anaphoric or not anaphoric ‘Right’ antecedent picked up
Measures developed for the competitions: Automatic way of doing the evaluation
More realistic measures (Byron, Mitkov) Accuracy on ‘hard’ cases (e.g., ambiguous
pronouns)
Vilain et al 1995
The official MUC scorer Based on precision and recall of links
Vilain et al: the goalThe problem: given that A,B,C and D are part of a coreference chain in the KEY, treat as equivalent the two responses:
And as superior to:
Vilain et al: RECALL
To measure RECALL, look at how each coreference chain Si in the KEY is partitioned in the RESPONSE, and count how many links would be required to recreate the original, then average across all coreference chains.
Vilain et al: Example recall
In the example above, we have one coreference chain of size 4 (|S| = 4)
The incorrect response partitions it in two sets (|p(S)| = 2)
R = 4-2 / 4-1 = 2/3
Vilain et al: precision
Count links that would have to be (incorrectly) added to the key to produce the response
I.e., ‘switch around’ key and response in the equation before
Beyond Vilain et al
Problems: Only gain points for links. No points
gained for correctly recognizing that a particular mention is not anaphoric
All errors are equal Proposals:
Bagga & Baldwin’s B-CUBED algorithm Luo recent proposal
Outline
A reminder: anaphora resolution, factors affecting the interpretation of anaphoric expressions
A brief history of anaphora resolution First algorithms: Charniak, Winograd, Wilks Pronouns: Hobbs Salience: S-List, LRC
Early ML work Definite descriptions: Vieira & Poesio The MUC initiative – also: coreference evaluation
methods Soon et al
Soon et al 2001
First ‘modern’ ML approach to anaphora resolution Resolves ALL anaphors Fully automatic mention identification
Developed instance generation & decoding methods used in a lot of work since
Soon et al: preprocessing
POS tagger: HMM-based 96% accuracy
Noun phrase identification module HMM-based Can identify correctly around 85% of mentions (??
90% ??) NER: reimplementation of Bikel Schwartz and
Weischedel 1999 HMM based 88.9% accuracy
Soon et al: training instances
<ANAPHOR (j), ANTECEDENT (i)>
Soon et al 2001: Features
NP type Distance Agreement Semantic class
Soon et al: NP type and distance
NP type of antecedent i i-pronoun (bool)
NP type of anaphor j (3) j-pronoun, def-np, dem-np (bool)
DIST 0, 1, ….
Types of both both-proper-name (bool)
Soon et al features: string match, agreement, syntactic position
STR_MATCHALIAS dates (1/8 – January 8) person (Bent Simpson / Mr. Simpson) organizations: acronym match (Hewlett Packard / HP)
AGREEMENT FEATURES number agreement gender agreement
SYNTACTIC PROPERTIES OF ANAPHOR occurs in appositive contruction
Soon et al: semantic class agreement
PERSON
FEMALE MALE
OBJECT
DATEORGANIZATION
TIME MONEY PERCENT
LOCATION
SEMCLASS = true iff semclass(i) <= semclass(j) or viceversa
Soon et al: generating training instances
Marked antecedent used to create positive instance
All mentions between anaphor and marked antecedent used to create negative instances
Generating training instances
((Eastern Airlines) executives) notified ((union) leaders) that (the carrier) wishes to discuss (selective (wage) reductions) on (Feb 3)
POSITIVE
NEGATIVE
NEGATIVE
NEGATIVE
Soon et al: decoding
Right to left, consider each antecedent until classifier returns true
Soon et al: evaluation
MUC-6: P=67.3, R=58.6, F=62.6
MUC-7: P=65.5, R=56.1, F=60.4
Soon et al: evaluation
Subsequent developments
Different models of the task Different preprocessing techniques Using lexical / commonsense knowledge
(particularly semantic role labelling) Next lecture
Salience Anaphoricity detection Development of AR toolkits (GATE,
LingPipe, GUITAR)
Other models
Cardie & Wagstaff: coreference as (unsupervised) clustering Much lower performance
Ng and Cardie Yang ‘twin-candidate’ model
Ng and Cardie
2002: Changes to the model:
Positive: first NON PRONOMINAL Decoding: choose MOST HIGH PROBABILITY
Many more features: Many more string features Linguistic features (binding, etc)
Subsequently: Discourse new detection (see below)
Readings
Kehler’s chapter in Jurafsky & Martin Alternatively: Elango’s survey
http://pages.cs.wisc.edu/~apirak/cs/cs838/pradheep-survey.pdf
Hobbs J.R. 1978, “Resolving Pronoun References,” Lingua, Vol. 44, pp. 311-. 338. Also in Readings in Natural Language Processing,
Renata Vieira, Massimo Poesio, 2000. An Empirically-based System for Processing Definite Descriptions. Computational Linguistics 26(4): 539-593
W. M. Soon, H. T. Ng, and D. C. Y. Lim, 2001. A machine learning approach to coreference resolution of noun phrases. Computational Linguistics, 27(4):521--544,