Responses to help-desk email inquiries are often repetitive, sharing
  varying degrees of commonality.  In addition, a significant
  proportion of the responses are generic, containing a very low level
  of technical content. In this paper, we present a corpus-based
  approach for identifying common elements in help-desk responses and
  using them to construct a new response.  A help-desk domain is
  unique in that responses that contain even one incongruous sentence
  can alienate a user.  It is therefore not always possible to
  automatically generate a complete response, because personalization
  is often better handled by human operators.  Our system is designed
  to find and collate the generic portions of responses.  We have
  adapted multi-document summarization techniques, and developed a
  measure that predicts the completeness of a planned response, thus
  indicating when a fully automated response is possible.  Our
  evaluation shows that 14\% of the responses in our corpus can be
  represented by complete generic responses.