Some Foundational Linguistic Elements for QA Systems: an Application

1

• Some Foundational Linguistic Elements for QA
  Systems an Application
• to E-government Services

Farida Aouladomar JURIX 05
9/11/05
2
Objective

• ? problematics of procedural texts for answering
  procedural questions on the web
• governments increasingly offer online services to
  their citizens.
• ex How can I get the french citizenship?
• ? Semantics and structure of procedural text
• answers to procedural questions are structures
• where to unify questions and parts of the
  procedural texts
• give a complete and appropriate response

3
Framework

• ? Advanced Question Answering

? QA Systems for factoid questions
InferenceWEB (McGuinness, 04), JAVELIN (Nyberg
and al, 03), Mulder (Kwok and al, 00),
WEBCOOP (Benamara, 04)
? Non-factoid questions comparison,
procedural, opinion, causal, etc.
? How-questions within a cooperative
environment
4
How to get a passport ?
Search engine
Filtering 1
Annotation
Filtering 2
5
Overview

• ? Procedural questions
• ? Grammar of procedural discourse
• - syntactic structure
• - linguistic marks
• ? Questionability and responses

6
Procedural questions (1)

• ? Procedural questions questions
• introduced by  comment 
• - Comment vas-tu ?
• - Comment dit-on maison en espagnol ?
• - Comment est mort John?
• - Comment on mange le couscous au Maroc ?
• - Comment payer mon billet davion ?
• - Comment changer une roue ?
• - Comment créer une entreprise ?

7
Procedural questions (2)

• ?Other forms of procedural questions
• Forms in que faire  , quel être
  proposition
• Ex que faire pour obtenir un visa ?, quelles
  sont les
• démarches à effectuer pour obtenir un visa pour
  lInde?
• quelle est la procédure pour obtenir la
  nationalité belge ?

8
Procedural questions (3)

• The elliptical use of  comment  key words
• Ex trouver un avocat, déposer une plainte,
  inscription sur liste électorale
• Lexical inference
• Ex jai perdu mon passeport
• Questions expecting an answer of
• procedural type
• Ex est-il possible davoir la double nationalité
  ?

9
Overview

• ? Procedural questions
• ? Grammar of procedural discourse
• - syntactic structure
• - linguistic marks
• ? Questionability and responses

10
Procedural texts

• ? Sequences of instructions operating over a
  specific set of entities in order to reach a
  goal.
• ? Goals subgoals skeletal structure of
  procedural texts
• ? Group of texts cooking receipes, maintenance
  manuals, assembly notices, directions for use,
  teaching texts, medical notices, social behavior
  recommendations, advice texts, savoir-faire
  guides, itinerary guides, architectory plans,
  musical scales, legislation, court decisions,
• legal guidelines procedures, etc.
• ? Corpus identify textual elements of
  procedural texts for the answer process.

11
Discursive Structure - Corpus (1)
? Structure analysis of procedural texts -
Corpora analysis method based on how queries
on the web and on manual enrichment
Questions classification Nb of procedural texts from queries inventory Nb of added procedural texts Total number of texts in our corpus
Communication / advices / e-government 48 0 48
Technical domain computer science assembly texts 30 20 50
Health 3 6 9
Receipes 0 10 10
Rules 0 7 7
total 81 43 124
12
Discursive Structure Grammar (2)
Text ? title, (summary), (warning),
(pre-requisites), (picture) lt objective.
Summary ? title.
Pre-requisites ? list of objects, (instruction
sequences).
Objective ? goallt (warning), (picture),
(pre-requisites), instruction sequences /
objective
Instruction sequences ? instseq lt connector lt
instruction sequences / instseq.
13
Grammar (3)
Instruction sequence ? imperative linear
sequence / optional sequence / alternative
sequence / imperative co-temporal sequence.
Imperative linear sequence ? instruction lt
temporal mark, imperative linear sequence /
instruction.
Optional sequence ? optionality expression,
imperative linear sequence.
Instruction ? (iterative expression), action,
(goal) (argument), (reference),(picture),
(warning)
14
Grammar (3)
Alternative sequence ? (conditional expression),
(argument), imperative linear sequence,
(alternative-opposition mark) lt instseq /
(conditional expression, instseq).
Imperative co-temporal sequence ? imperative
linear sequence lt co-temporal mark lt
imperative co-temporal sequence / instruction.
Instruction ? (iterative expression), action,
(goal) (argument), (reference), (picture),
(warning)
15
Texte ? title, (summary), (warning),
(pre-requisites), (picture) lt objective.
16
Objective ? goallt (warning), (picture),
(pre-requisites), instruction sequences /
objective
17
Imperative linear sequence ? instruction lt
(temporal mark) lt imperative linear sequence /
instruction
Instruction ? (iterative expression), action,
(goal) (argument), (reference), (picture),
(warning)
La premiere étape consiste à ouvrir entièrement
le boîtier, puis de le placer à plat sur une
surface large où vous aurez suffisamment de
place pour travailler confortablement, et
enfin retirer tous les caches en plastiques des
baies à lavant du PC
La première étape consiste à ouvrir entièrement
le boîtier, puis de le placer à plat sur une
surface large où vous aurez suffisamment de
place pour travailler confortablement, et
enfin retirer tous les caches en plastiques des
baies à lavant du PC
Temporal marks
The first stage consists in fully-opening the
box, then place it on a large surface where you
will have sufficient place to work
comfortably, and finally to withdraw the plastic
protections on the PC front side.
18
Texte ? title, (summary), (warning),
(pre-requisites), (picture) lt objective.
19
annotation

• lt objectivegt ltgoalgt Postez le formulaire et les
  documents au Centre de traitement des demandes.
  lt\goalgt
• lt instseqgt lt imper_lineargt lttemp_markgt Après
  lt\temp_markgt ltinstrgt avoir rempli le formulaire
  de demande, lt\instrgt ltinstrgt vous devez le
  poster, dans lenveloppe-réponse fournie, à
  ladresse suivante 
• Centre de traitement des demandesCitoyenneté et
  Immigration CanadaC.P. 7000Sydney
  (Nouvelle-Écosse) B1P 6V6 lt\instrgt lt
  imper_lineargt lt \instseqgt
• ltwarninggt Noubliez pas
• de signer et de dater le formulaire ainsi que de
  signer vos photos
• dinclure le reçu du paiement (formulaire
  IMM 5401)
• de mettre votre demande dans lenveloppe
• de mettre les photos dans lenveloppe
• dinclure les photocopies de tous les documents
  requis. ltwarninggt
• lt\objectivegt

lt instseqgt lt imper_lineargt lttemp_markgt Après
lt\temp_markgt ltinstrgt avoir rempli le formulaire
de demande, lt\instrgt ltinstrgt vous devez le
poster, dans lenveloppe-réponse fournie, à
ladresse suivante  Centre de traitement des
demandesCitoyenneté et Immigration CanadaC.P.
7000Sydney (Nouvelle-Écosse) B1P 6V6 lt\instrgt
lt\imper_lineargt lt \instseqgt

• ltwarninggt Noubliez pas
• ltinstrgt de signer et de dater le formulaire ainsi
  que de signer vos photos lt\instrgt
• ltinstrgt dinclure le reçu du paiement (formulaire
  IMM 5401) lt\instrgt
• ltinstrgt de mettre votre demande dans lenveloppe
  lt\instrgt
• ltinstrgt de mettre les photos dans lenveloppe
  lt\instrgt
• ltinstrgt dinclure les photocopies de tous les
  documents requis. lt\instrgt ltwarninggt

20
Grammar limits

• The grammar is
• -- too rigid, and not sufficient to describe all
  facets of procedural texts
• -- not very constrained
• -- not explicative and predictive enough
• A new orientation of the discursive structure of
  procedural texts principles and norms

21
Overview

• ? Procedural questions
• ? Grammar of procedural discourse
• - syntactic structure
• - linguistic marks
• ? Questionability and responses

22
Linguistic marks (1)

• ? Discursive marks allow for the identification
  of the elements of the grammar
• -- classical temporal marks (precedence,
  overlap, inclusion, etc.)
• -- restrictions, conditions, alternatives,
  comparisons, etc.
• -- causal marks (identification of objectives,
  goals, warnings, preventions, consequences, etc.)

23
instruction localization (1)
-- typoraphic criterion
24
instruction localization (2)
-- morphological criterion (ex imperatives)
25
instruction localization (3)
-- semantic criterion action verbs
26
Overview

• ? Procedural questions
• ? Grammar of procedural discourse
• - syntactic structure
• - linguistic marks
• ? Questionability and responses

27
(No Transcript)
28
Questionability (1)

• ? The ability or the relevance of any text to
  respond to How-questions.
• ? Stage 1 determine the  CATEG  rate
• -- measure the procedural nature of a text
• ? Stage 2 determine the  QUEST  rate
• -- measure of zones which are qualified for
  potential question unification

29
Questionability (2)

• ? Stage 1  CATEG  gt 3  surface criteria 
• - typographic forms (TF)
• - morpho-syntactic marks (MSM)
• - articulatory marks (AM) (temporal,
  argumentative, etc.)
• ? average frequency is computed for each criteria
  for all texts (noted as TFaverage, MSMaverage,
  AMaverage)

30
Questionability (3)

• ? For text i, we Define its CATEG rate

? Selection of the  best  texts gt annotation
31
Questionability (4)

• ? Stage 2  Quest  gt evaluate the number of
  areas which can potentially match with
  How-questions
• ? 4 areas leading to 4 criteria
• - number of titles (TIT)
• - action verbs (AV)
• - number of goals (GOA)
• - manners (MAN)

32
Questionability (5)
? For text i, we define its QUEST rate 
? Question Unification
33
Responses
? 3 tasks -- selecting procedural texts which
have the best questionabilty rate -- matching
the question body with  questionable
zones  -- extracting the relevant portion of
the text and returning it to the user in a
user-friendly way
34
To Conclude
? Still experimental work - ongoing design of
a system that annotates texts and evaluates the
questionability rate ? Perspectives -
validation / evaluation - unification ? give a
complete answer to the user - respond
cooperatively to procedural questions ? Work
more specifically on the structure of legal /
administrative procedural texts
35
(No Transcript)