What matters in a transferable neural network model for relation classification in the biomedical domain?

A lack of sufficient labeled data often limits the applicability of advanced machine learning algorithms to real life problems. However, the efficient use of transfer learning (TL) has been shown to be very useful across domains. TL make use of valuable knowledge learned in one task (source task), w...

Descripción completa

Detalles Bibliográficos
Publicado en:Artificial Intelligence in Medicine Vol. 87; pp. 60 - 67
Autores principales: Sahu, Sunil Kumar, Anand, Ashish
Formato: Journal Article
Publicado: Elsevier B.V. May2018
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=129825965&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 129825965
    longDbName: CINAHL Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    dissinfo:
    jinfo:
      jid:
        09333657
        3HY
      jtl: Artificial Intelligence in Medicine
      issn: 09333657
      maglogo: N
    pubinfo:
      dt: May2018
      vid: 87
      pid: 1004
      pub: Elsevier B.V.
    artinfo:
      ui:
        129825965
        129825965
        NLM29661577
        10.1016/j.artmed.2018.03.006
        NLM29661577
        129825965
      ppf: 60
      ppct: 7
      formats:
      tig:
        atl: What matters in a transferable neural network model for relation classification in the biomedical domain?
      aug:
        au:
          Sahu, Sunil Kumar
          Anand, Ashish
        affil: Department of Computer Science and Engineering, Indian Institute of Technology Guwahati, India
      sug:
        subj:
          Neural Networks (Computer)
          Data Collection
          Algorithms
          Drug Interactions
          Questionnaires
          Scales
      ab: A lack of sufficient labeled data often limits the applicability of advanced machine learning algorithms to real life problems. However, the efficient use of transfer learning (TL) has been shown to be very useful across domains. TL make use of valuable knowledge learned in one task (source task), where sufficient data is available, in order to improve performance on the task of interest (target task). In the biomedical and clinical domain, a lack of sufficient training data means that machine learning models cannot be fully exploited. In this work, we present two unified recurrent neural models leading to three transfer learning frameworks for relation classification tasks. We systematically investigate the effectiveness of the proposed frameworks in transferring knowledge from a source task to a target task when the characteristics of the source data vary, such as similarity or relatedness between the source and target tasks, and the size of training data for the source task. Our empirical results show that the proposed frameworks, in general, improve the model performance. However, these improvements do depend on characteristics of source and target tasks. This dependence then finally determine the choice of a particular TL framework.
      pubtype: Academic Journal
      doctype: Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N