Conference paper Open Access

A Case Study of Closed-Domain Response Suggestion with Limited Training Data

Galke, Lukas; Gerstenkorn, Gunnar; Scherp, Ansgar


MARC21 XML Export

<?xml version='1.0' encoding='UTF-8'?>
<record xmlns="http://www.loc.gov/MARC21/slim">
  <leader>00000nam##2200000uu#4500</leader>
  <datafield tag="041" ind1=" " ind2=" ">
    <subfield code="a">eng</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">conversational agents</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">neural networks</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">representation learning</subfield>
  </datafield>
  <controlfield tag="005">20191114190956.0</controlfield>
  <datafield tag="500" ind1=" " ind2=" ">
    <subfield code="a">This is a post-peer-review, pre-copyedit version of a paper published in Elloumi M, Granitzer M, Hameurlain A, Seifert C, Stein
B, Tjoa A &amp; Wagner R (eds.) Database and Expert Systems Applications. DEXA 2018. Communications in Computer and
Information Science, 903. The final authenticated version is available online at: https://doi.org/10.1007/978-3-319-99133-7_18</subfield>
  </datafield>
  <controlfield tag="001">2583130</controlfield>
  <datafield tag="711" ind1=" " ind2=" ">
    <subfield code="d">3-6 September 2018</subfield>
    <subfield code="g">DEXA 2018</subfield>
    <subfield code="a">Database and Expert Systems Applications - DEXA 2018 International Workshops, BDMICS, BIOKDD, and TIR</subfield>
    <subfield code="c">Regensburg, Germany</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">University of Potsdam</subfield>
    <subfield code="0">(orcid)0000-0002-4889-511X</subfield>
    <subfield code="a">Gerstenkorn, Gunnar</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">University of Stirling</subfield>
    <subfield code="0">(orcid)0000-0002-2653-9245</subfield>
    <subfield code="a">Scherp, Ansgar</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">235300</subfield>
    <subfield code="z">md5:571bc5ceec1bb2e130ebd00a6a56c293</subfield>
    <subfield code="u">https://zenodo.org/record/2583130/files/A_Case_Study_of_Closed_Domain_Response_Suggestion_with_Limited_Training_Data.pdf</subfield>
  </datafield>
  <datafield tag="542" ind1=" " ind2=" ">
    <subfield code="l">open</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="y">Conference website</subfield>
    <subfield code="u">http://www.dexa.org/</subfield>
  </datafield>
  <datafield tag="260" ind1=" " ind2=" ">
    <subfield code="c">2018-09-06</subfield>
  </datafield>
  <datafield tag="909" ind1="C" ind2="O">
    <subfield code="p">user-moving-h2020</subfield>
    <subfield code="o">oai:zenodo.org:2583130</subfield>
  </datafield>
  <datafield tag="100" ind1=" " ind2=" ">
    <subfield code="u">Kiel University</subfield>
    <subfield code="0">(orcid)0000-0001-6124-1092</subfield>
    <subfield code="a">Galke, Lukas</subfield>
  </datafield>
  <datafield tag="245" ind1=" " ind2=" ">
    <subfield code="a">A Case Study of Closed-Domain Response Suggestion with Limited Training Data</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">user-moving-h2020</subfield>
  </datafield>
  <datafield tag="536" ind1=" " ind2=" ">
    <subfield code="c">693092</subfield>
    <subfield code="a">Training towards a society of data-savvy information professionals to enable open leadership innovation</subfield>
  </datafield>
  <datafield tag="540" ind1=" " ind2=" ">
    <subfield code="u">http://creativecommons.org/licenses/by/4.0/legalcode</subfield>
    <subfield code="a">Creative Commons Attribution 4.0 International</subfield>
  </datafield>
  <datafield tag="650" ind1="1" ind2="7">
    <subfield code="a">cc-by</subfield>
    <subfield code="2">opendefinition.org</subfield>
  </datafield>
  <datafield tag="520" ind1=" " ind2=" ">
    <subfield code="a">&lt;p&gt;We analyze the problem of response suggestion in a closed domain along a real-world scenario of a digital library. We present a text-processing pipeline to generate question-answer pairs from chat transcripts. On this limited amount of training data, we compare retrieval-based, conditioned-generation, and dedicated representation learning approaches for response suggestion. Our results show that retrieval-based methods that strive to find similar, known contexts are preferable over parametric approaches from the conditioned-generation family, when the training data is limited. We, however, identify a specific representation learning approach that is competitive to the retrieval-based approaches despite the training data limitation.&lt;/p&gt;</subfield>
  </datafield>
  <datafield tag="024" ind1=" " ind2=" ">
    <subfield code="a">10.1007/978-3-319-99133-7_18</subfield>
    <subfield code="2">doi</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">publication</subfield>
    <subfield code="b">conferencepaper</subfield>
  </datafield>
</record>
20
22
views
downloads
Views 20
Downloads 22
Data volume 5.2 MB
Unique views 18
Unique downloads 20

Share

Cite as