Conference paper Open Access

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

Galke, Lukas; Gerstenkorn, Gunnar; Scherp, Ansgar


JSON Export

{
  "files": [
    {
      "links": {
        "self": "https://zenodo.org/api/files/f7d12c7e-9295-474c-a312-cb71d997c3c7/A_Case_Study_of_Closed_Domain_Response_Suggestion_with_Limited_Training_Data.pdf"
      }, 
      "checksum": "md5:571bc5ceec1bb2e130ebd00a6a56c293", 
      "bucket": "f7d12c7e-9295-474c-a312-cb71d997c3c7", 
      "key": "A_Case_Study_of_Closed_Domain_Response_Suggestion_with_Limited_Training_Data.pdf", 
      "type": "pdf", 
      "size": 235300
    }
  ], 
  "owners": [
    40585
  ], 
  "doi": "10.1007/978-3-319-99133-7_18", 
  "stats": {
    "version_unique_downloads": 18.0, 
    "unique_views": 15.0, 
    "views": 17.0, 
    "downloads": 20.0, 
    "unique_downloads": 18.0, 
    "version_unique_views": 15.0, 
    "volume": 4706000.0, 
    "version_downloads": 20.0, 
    "version_views": 17.0, 
    "version_volume": 4706000.0
  }, 
  "links": {
    "doi": "https://doi.org/10.1007/978-3-319-99133-7_18", 
    "latest_html": "https://zenodo.org/record/2583130", 
    "bucket": "https://zenodo.org/api/files/f7d12c7e-9295-474c-a312-cb71d997c3c7", 
    "badge": "https://zenodo.org/badge/doi/10.1007/978-3-319-99133-7_18.svg", 
    "html": "https://zenodo.org/record/2583130", 
    "latest": "https://zenodo.org/api/records/2583130"
  }, 
  "created": "2019-03-04T16:55:49.445496+00:00", 
  "updated": "2019-03-07T11:11:04.502826+00:00", 
  "conceptrecid": "2583129", 
  "revision": 3, 
  "id": 2583130, 
  "metadata": {
    "access_right_category": "success", 
    "doi": "10.1007/978-3-319-99133-7_18", 
    "description": "<p>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.</p>", 
    "license": {
      "id": "CC-BY-4.0"
    }, 
    "title": "A Case Study of Closed-Domain Response Suggestion with Limited Training Data", 
    "language": "eng", 
    "notes": "This is a post-peer-review, pre-copyedit version of a paper published in Elloumi M, Granitzer M, Hameurlain A, Seifert C, Stein\nB, Tjoa A & Wagner R (eds.) Database and Expert Systems Applications. DEXA 2018. Communications in Computer and\nInformation Science, 903. The final authenticated version is available online at: https://doi.org/10.1007/978-3-319-99133-7_18", 
    "relations": {
      "version": [
        {
          "count": 1, 
          "index": 0, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "2583129"
          }, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "2583130"
          }
        }
      ]
    }, 
    "communities": [
      {
        "id": "moving-h2020"
      }
    ], 
    "grants": [
      {
        "code": "693092", 
        "links": {
          "self": "https://zenodo.org/api/grants/10.13039/501100000780::693092"
        }, 
        "title": "Training towards a society of data-savvy information professionals to enable open leadership innovation", 
        "acronym": "MOVING", 
        "program": "H2020", 
        "funder": {
          "doi": "10.13039/501100000780", 
          "acronyms": [
            "EC"
          ], 
          "name": "European Commission", 
          "links": {
            "self": "https://zenodo.org/api/funders/10.13039/501100000780"
          }
        }
      }
    ], 
    "keywords": [
      "conversational agents", 
      "neural networks", 
      "representation learning"
    ], 
    "publication_date": "2018-09-06", 
    "creators": [
      {
        "orcid": "0000-0001-6124-1092", 
        "affiliation": "Kiel University", 
        "name": "Galke, Lukas"
      }, 
      {
        "orcid": "0000-0002-4889-511X", 
        "affiliation": "University of Potsdam", 
        "name": "Gerstenkorn, Gunnar"
      }, 
      {
        "orcid": "0000-0002-2653-9245", 
        "affiliation": "University of Stirling", 
        "name": "Scherp, Ansgar"
      }
    ], 
    "meeting": {
      "acronym": "DEXA 2018", 
      "url": "http://www.dexa.org/", 
      "dates": "3-6 September 2018", 
      "place": "Regensburg, Germany", 
      "title": "Database and Expert Systems Applications - DEXA 2018 International Workshops, BDMICS, BIOKDD, and TIR"
    }, 
    "access_right": "open", 
    "resource_type": {
      "subtype": "conferencepaper", 
      "type": "publication", 
      "title": "Conference paper"
    }
  }
}
17
20
views
downloads
Views 17
Downloads 20
Data volume 4.7 MB
Unique views 15
Unique downloads 18

Share

Cite as