Extracting Information from Participial Structures

1
Extracting Information from Participial Structures

• Kata Gábor, Eniko Héja, Ágnes Mészáros
• Research Institute for Linguistics, HAS
• 8th INTEX WORKSHOP, 2005

2
STRUCTURE

• IE system and its shortage the problem of
  participles
• NPs and participles in Hungarian
• a possible enhancement of the IE system
• implementation in INTEX

3
IE system

• input text (1-2 sentences of short business news)
• shallow syntactic analysis
• pre-defined semantic patterns (event frames)
• output event frames slots filled by the
  elements of the input text
• the event, its participants and circumstances are
  identified

4
Event frames

• Az ABN Amro Bank egyesül a Kereskedelmi és
  Hitelbankkal.
• ABN Amro Bank fuses with Commercial and Credit
  Bank.
• ltevent schema"owner_changed.fusion.6"
  roles_matched"3/3"gt
• ltrv role"member_company_1" pos"N" case"NOM"
  sem"companyinstitute"gt
• ltNP id"88" sem"company countable human
  institute"gt
• ltw id"0" class"DET" at"1-1" lex"az"
  case"NOM"gtAzlt/wgt
• ltw id"2" class"UNKNOWN" at"2-2"
  lex"ABN"gtABNlt/wgt
• ltw id"4" class"UNKNOWN" at"3-3"
  lex"Amro"gtAmrolt/wgt
• ltw id"6" class"N" at"4-4" lex"bank"
  case"NOM"gtBanklt/wgt
• lt/NPgt
• lt/rvgt
• ltrv role"_1" pos"V" lemma"egyesül"gt
• ltw id"8" class"V" at"5-5" lex"egyesül"gtegyesü
  llt/wgt
• lt/rvgt
• ltrv role"member_company_2" pos"N" case"INS"
  sem"companyinstitute"gt
• ltNP id"118" sem"company countable institute"gt
• ltw id"13" class"DET" at"6-6" lex"a"
  case"NOM"gtalt/wgt

5
Mapping syntax to event frames

• SYNTAX EVENT FRAMES
• verb main event
• arguments participants
• free modifiers circumstances
• (time,
  location,manner...)

6
Mapping syntax to event frames

• Problem secondary information (cause or
  antecedent of the main event) is hidden in
  participial structures
• A befektetok által tegnap eladott részvények
  megnövelték a tozsde forgalmát.
• The shares sold yesterday by the investors
  increased the traffic at the stock exchange.

7
Mapping syntax to event frames

• A befektetok által tegnap eladott részvények
  megnövelték a tozsde forgalmát.
• The shares sold yesterday by the investors
  increased the traffic at the stock exchange.

részvények / shares /
a befektetok / the investors /
eladott / sold /
tegnap / yesterday /
8
Mapping syntax to event frames

• A befektetok által tegnap eladott részvények
  megnövelték a tozsde forgalmát.
• The shares sold yesterday by the investors
  increased the traffic at the stock exchange.
• A befektetok tegnap eladtak részvényeket.
• The investors sold shares yesterday.

9
A solution

• a preprocessing module within the IE system which
  transforms participial structures into sentences
  with a finite predicate
• semantic frame matching may operate on
  transformed sentences
• 1st step past participles within NPs
• the participle preserves the meaning of its base
  verb
• its arguments can be derived from the internal
  structure of the NP

10
NPs in Hungarian 1.
11
NPs in Hungarian 2.

• DET head Noun

ADV
NPcase
Participles (past, present)
modifiers
APcase
NPostp
V.INF
...
12
Participles in Hungarian

• ADJ Participle homonimy is a problem
• mérsékelt PC-chip kereslet
• modest /moderated/ demand for PC-chips
• Valaki mérsékelte a PC-chip keresletet.
• Somebody moderated the demand for PC-chips
• ragozott szóalakok
• inflected word forms
• Valaki ragozott szóalakokat.
• Somebody inflected word forms.
• only participles can be transformed

13
Participle or Adjective?

• syntactic tests
• comparative
• ADV formation
• predicative use
• impossibility of preverb detachment
• we need to decide in the context whether the
  given word form is an ADJ or a PART
• 1. If at least one of the base verbs
  complements is present, than it is a participle.

14
Participle or Adjective?

• syntactic tests
• comparative
• ADV formation
• predicative use
• preverb detachment
• we need to decide in the context whether the
  given word form is an ADJ or a PART
• 2. If at least one of the base verbs
  complements / adjuntcs / a preverb is present,
  than it is a participle.

15
Participle or Adjective?

• TESTS
• comparative mérsékeltebb kereslet
• more moderate demand
• predicative Ez a szóalak ragozott.
• This word form is
  inflected.
• ADV formation mérsékelt ? mérsékelten
• moderate ? moderately
• preverb detachment
• a fel nem újított házak ?
• the re- not stored houses (not
  restored)
• Ezek a házak fel nem újítottak.
• These houses are re- not stored.

16
THE GRAMMAR

• - the correctness and informativity of the
  resulting sentence depends on the correct
  identification of verbal arguments and modifiers
  within the NP
• - then these elements are transformed according
  to their grammatical function
• past participles may be formed from both
  transitive or intransitive verbs
• if the base verb is intransitive, the head noun
  of the NP represents the subject of the base
  verb
• az összedolt épület /the collapsed building/
• if the base verb is transitive, the head noun
  represents the direct object of the base verb
• a bejelentett változások /the changes
  announced/
• ? transitivity needs to be coded

17
THE GRAMMAR

• transformation rules are (enhanced) FSTs
• they store relevant elements of the input NP in
  variables
• the output is made up of the content of these
  variables but in an altered order function
  words needed in the sentence
• our delaf dictionary codes
• transitivity properties of verbs (on the basis of
  a lexicon-grammar of verbal argument structures)
• - preverb feature shows whether the base verb
  has a preverb

18
Transformation Graphs 1.Transitive Verbs

• transitive verbs without expressed subject
• (somebody insertion)
• Det (V_compl) VMIB
  N
• ? Valaki V_vmib Det N t (V_compl) .
• transitive verbs with a subject with the PostP
  által
• Det Nsubj által (V_compl)
  VMIB N
• ? Nsubj V_vmib Det N t
  (V_compl) .

19
Transformation Graphs 2.Intransitive Verbs

• head N becomes subject (patient)
• Det (V_compl) VMIB N
• ? Det N V_vmib (V_compl) .

20
Structure of the graphs

• 1 graph
• 3 subgraphs according to complement-types
  possessor / verbal complementadjunct / nothing/
• each subgraph divided into two paths
• transitive / intransitive verbs

21
(No Transcript)
22
(No Transcript)
23
Evaluation

• central aspect to what extent does it augment
  the efficiency of the IE system?
• lack of information (recall value) is considered
  less important than incorrect information
  (precision)
• evaluated on the 231.000 words corpus of short
  business news
• 1259 hits ? 898 qualified as informative
• precision 64
• further task recall
• (requires a corpus with manually annotated
  participial structures)

24
THANK YOU FOR YOUR ATTENTION!
gkata, eheja, magnes_at_corpus.nytud.hu