Parsing
Chunk Chunk Parsing
Also called chunking, light parsing, or partial parsing. Method: Assign some additional structure to input over tagging Used when full parsing not feasible or not desirable. Because of the expense of full-parsing, often treated as a stop-gap solution.
Chunk Parsing
No rich hierarchy, as in parsing. Usually one layer above tagging.
The process:
1. Tokenize 2. Tag 3. Chunk
Chunk Parsing
Like tokenizing and tagging in a few respects:
1. Can skip over material in the input
1. Often finite-state (or finite-state like) methods are used (applied over tags) 2. Often application specific (i.e., the chunks tagged have uses for particular applications)
Chunk Parsing
Chief Motivations: to find data or to ignore data Example from Bird and Loper: find the argument structures for the verb give. Can "discover" significant grammatical structures before developing a grammar: gave NP gave up NP in NP gave NP up gave NP help
Chunk Parsing
Like parsing, except:
It is not exhaustive, and doesn't pretend to be.
Structures and data can be skipped when not convenient or not desired Nested structures typical in parsing
Structures of fixed depth produced
[S[NP The cow [PP in [NP the barn]]] ate
Not in chunking
[NP The cow] in [NP the barn] ate
Chunk Parsing
Finds contiguous, non-overlapping spans of related text, and groups them into chunks. Because contiguity is given, finite state methods can be adapted to chunking
Longest Match
Abney 1995 discusses longest match heuristic:
One automaton for each phrasal category Start automata at position i (where i=0 initially) Winner is the automaton with the longest match
Longest Match
He took chunks from the PTB: NP D N NP D Adj N VP V Encoded each rule as an automaton Stored longest matching pattern (the winner) If no match for a given word, skipped it (in other words, didn't chunk it) Results: Precision .92, Recall .88
An Application
Data-Driven Linguistics Ontology Development (NSF BCE-0411348) One focus: locate linguistically annotated (read: tagged) text and extract linguistically relevant terms from text Attempt to discover "meaning" of the terms Intended to build out content of the ontology (GOLD) Focus on Interlinear Glossed Text (IGT)
An Application
Interlinear Glossed Text (IGT), some examples:
(1) Afisi
a-na-ph-a
nsomba (Baker 1988:254)
hyenas SP-PST-kill-ASP fish `The hyenas killed the fish.'
An Application
More examples:
(4) a. yerexa-n p'at'uhan-e bats-ets child-NOM window-ACC openAOR.3SG `The child opened the window.'
(Megerdoomian ??)
(37)
t3ile-3u33i p3-s 1pl-TOP buy-NOM 'What we to buy is meat.' need is meat.'
pies meat
3u3. COP (LaPolla 2000)
An Application
(4) a. yerexa-n p'at'uhan-e bats-ets child-NOM window-ACC open-AOR.3SG `The child opened the window.' (Megerdoomian ??)
Problem: How do we `discover' the meaning of the linguistically salient terms, such as NOM, ACC, AOR, 3SG? Perhaps we can discover the meanings by examining the contexts in which the occur. POS can be a context. Problem: POS tags rarely used in IGT How do you assign POS tags to a language you know nothing about? IGT gives us aligned text for free!!
An Application
(4) a. yerexa-n p'at'uhan-e bats-ets window-ACC child-NOM open-AOR.3SG `The child opened the window.' (Megerdoomian ??) DT NN VBP DT NN
IGT gives us aligned text for free!! POS tag the English translation Align with the glosses and language data That helps. We now know that NOM and ACC attach to nouns, not verbs (nominal inflections) And AOR and 3SG attach to verbs (verbal inflections)
An Application
(4) a. yerexa-n p'at'uhan-e bats-ets child-NOM window-ACC open-AOR.3SG `The child opened the window.' (Megerdoomian ??) DT NN VBP DT NN
In the LaPolla example, we know that NOM does not attach to nouns, but to verbs. Must be some other kind of NOM.
(37) t3ile-3u33i p3-s 1pl-TOP buy-NOM 'What we to buy is meat.' need is meat.' pies meat 3u3. COP (LaPolla 2000)
An Application
(4) a. yerexa-n p'at'uhan-e bats-ets child-NOM window-ACC open-AOR.3SG `The child opened the window.' (Megerdoomian ??) DT NN VBP DT NN
How we tagged:
Globally applied most frequent tags (stupid tagger) Repaired tags where context dictated a change (e.g., TO preceding race = VB) Technique similar to Brill 1995
An Application
(4) a. yerexa-n p'at'uhan-e bats-ets child-NOM window-ACC open-AOR.3SG `The child opened the window.' (Megerdoomian ??) DT NN VBP DT NN
But can we get more information about NOM, ACC, etc.? Can chunking tell us something more about these terms? Yes!
An Application
(4) a. yerexa-n p'at'uhan-e bats-ets child-NOM window-ACC open-AOR.3SG `The child opened the window.' (Megerdoomian ??) DT NN VBP DT NN
Chunk phrases, mainly NPs Since relationship (in simple sentences) between NPs and verbs tells us something about the verbs' arguments (Bird and Loper 2005)... We can tap this information to discover more about the linguistic tags
An Application
(4) a. yerexa-n p'at'uhan-e bats-ets child-NOM window-ACC open-AOR.3SG `The child opened the window.' (Megerdoomian ??) DT NN VBP DT NN NP VP NP
Apply Abney 1995's longest match heuristic to get as many chunks as possible (especially NP) Leverage English canonical SVO (NVN) order to identify simple argument structures Use these to discover more information about the terms Thus...
An Application
(4) a. yerexa-n p'at'uhan-e bats-ets child-NOM window-ACC open-AOR.3SG `The child opened the window.' (Megerdoomian ??) DT NN VBP DT NN NP VP NP
We know that
NOM attaches to subject NPs may be a case marker indicating subject ACC attaches to object NPs may be a case marker indicating object
An Application
What we do next: look at co-occurrence relations (clustering) of
Terms with terms Host categories with terms
To determine more information about the terms Done by building feature vectors of the various linguistic grammatical terms ("grams") representing their contexts And measuring relative distances between these vectors (in particular, for terms we know)
Linguistic "Gram" Space
PASS CAUS PRES FUT PAST F 1SG M 3SG DAT GEN NOM ACC PL INSTR POSS LOC ABS ERG 18 19 13 14 12 11 17 10 16 3 4 1 2 15 8 9 5 7 6 P P !H~ + P !H~ + P H ! + P H ! + P P "H Q* P V "H Q* P P !H~ + # Q * P !H~ + # Q * + H ! AGREEMENT STD CASES H+ D P !H~ + # Q * P P ! + H P OBL CASES NOMINAL # Q * V P ! + H # Q * TNS/ASPECT " H Q* PH"Q* +H D VERBAL @ P P P P P P P P