@misc{GalkeGerstenkornScherp2018,
  author    = {Galke, Lukas and Gerstenkorn, Gunnar and Scherp, Ansgar},
  title     = {A case atudy of closed-domain response suggestion with limited training data},
  series = {Database and Expert Systems Applications : DEXA 2018 Iinternational workshops},
  volume    = {903},
  journal   = {Database and Expert Systems Applications : DEXA 2018 Iinternational workshops},
  publisher = {Springer},
  address   = {Berlin},
  isbn      = {978-3-319-99133-7},
  issn      = {1865-0929},
  doi       = {10.1007/978-3-319-99133-7_18},
  pages     = {218 -- 229},
  year      = {2018},
  abstract  = {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.},
  language  = {en}
}