Question Classification Transfer
Ligozat, Anne-Laure

Article Structure

Abstract

Question answering systems have been developed for many languages, but most resources were created for English, which can be a problem when developing a system in another language such as French.

Introduction

In question answering (QA), as in most Natural Language Processing domains, English is the best resourced language, in terms of corpora, lexicons, or systems.

Problem definition

A Question Answering (QA) system aims at returning a precise answer to a natural language question: if asked ”How large is the Lincoln Memorial?”, a QA system should return the answer ”164 acres” as well as a justifying snippet.

Transferng question classification

The two methods tested for transfering the classification, following (J abaian et al., 2011), are presented in Figure l:

Experiments

4.1 Question classes

Related work

Most question answering systems include question classification, which is generally based on supervised learning.

Conclusion

This paper presents a comparison between two transfer modes to adapt question classification from English to French.

Topics

question answering

Appears in 6 sentences as: Question Answering (2) Question answering (1) question answering (3)
In Question Classification Transfer
  1. Question answering systems have been developed for many languages, but most resources were created for English, which can be a problem when developing a system in another language such as French.
    Page 1, “Abstract”
  2. In question answering (QA), as in most Natural Language Processing domains, English is the best resourced language, in terms of corpora, lexicons, or systems.
    Page 1, “Introduction”
  3. While developing a question answering system for French, we were thus limited by the lack of resources for this language.
    Page 1, “Introduction”
  4. Section 5 details the related works in Question Answering .
    Page 1, “Introduction”
  5. A Question Answering (QA) system aims at returning a precise answer to a natural language question: if asked ”How large is the Lincoln Memorial?”, a QA system should return the answer ”164 acres” as well as a justifying snippet.
    Page 1, “Problem definition”
  6. Most question answering systems include question classification, which is generally based on supervised learning.
    Page 4, “Related work”

See all papers in Proc. ACL 2013 that mention question answering.

See all papers in Proc. ACL that mention question answering.

Back to top.

machine translation

Appears in 4 sentences as: machine translation (4)
In Question Classification Transfer
  1. We thus investigated the possibility of using machine translation to create a parallel corpus, as has been done for spoken
    Page 1, “Introduction”
  2. The idea is that using machine translation would enable us to have a large training corpus, either by using the English one and translating the test corpus, or by translating the training corpus.
    Page 1, “Introduction”
  3. One of the questions posed was whether the quality of present machine translation systems would enable to learn the classification properly.
    Page 1, “Introduction”
  4. well handled by all machine translation systems 2.
    Page 2, “Experiments”

See all papers in Proc. ACL 2013 that mention machine translation.

See all papers in Proc. ACL that mention machine translation.

Back to top.

n-grams

Appears in 4 sentences as: n-grams (4)
In Question Classification Transfer
  1. Table 1: Question classification precision for both levels of the hierarchy (features = word n-grams , classifier = libsvm)
    Page 3, “Experiments”
  2. Using word n-grams , monolingual English classification obtains .798 correct classification for the fine grained classes, and .90 for the coarse grained classes, results which are very close to those obtained by (Zhang and Lee, 2003).
    Page 3, “Experiments”
  3. Table 2: Question classification precision for both levels of the hierarchy (features = word n-grams with abbreviations, classifier = libsvm)
    Page 3, “Experiments”
  4. Table 3: Question classification precision for both levels of the hierarchy (features = word n-grams with abbreviations, classifier = libsvm)
    Page 4, “Experiments”

See all papers in Proc. ACL 2013 that mention n-grams.

See all papers in Proc. ACL that mention n-grams.

Back to top.

translation systems

Appears in 3 sentences as: translation systems (3)
In Question Classification Transfer
  1. One of the questions posed was whether the quality of present machine translation systems would enable to learn the classification properly.
    Page 1, “Introduction”
  2. well handled by all machine translation systems 2.
    Page 2, “Experiments”
  3. 2We tested other translation systems , but Google Translate gave the best results.
    Page 2, “Experiments”

See all papers in Proc. ACL 2013 that mention translation systems.

See all papers in Proc. ACL that mention translation systems.

Back to top.