Skip to main navigation Skip to search Skip to main content

Email answering by matching question and context-specific text patterns: Performance and error analysis

  • Eriks Sneiders
  • , Jonas Sjöbergh
  • , Alyaa Alfalahi
  • Stockholm University
  • Hokkaido University

Research output: Chapter in Book/Report/Conference proceedingConference paperResearchpeer-review

3 Citations (Scopus)

Abstract

Automated answering of frequent email inquiries is a text categorization task with narrow text categories, where all messages in one text category have the same answer. Such email categorization is optimized for high precision and at least acceptable recall. We tested matching of surface text patterns to nearly ten thousand email messages and achieved around 90% precision; the corresponding recall figures were 45-75% in different text categories. In order to achieve this performance level, the text patterns are designed to identify both the context of an email inquiry and the actual need that has created the inquiry – a question, request, or complaint. Our error analysis has pinpointed 12 reasons why text pattern matching may fail.

Original languageEnglish
Title of host publicationNew Advances in Information Systems and Technologies
EditorsMarcelo Mendonca Teixeira, Ana Maria Correia, Hojjat Adeli, Alvaro Rocha, Luis Paulo Reis
PublisherSpringer Verlag
Pages123-133
Number of pages11
ISBN (Print)9783319312316
DOIs
Publication statusPublished - 2016
Externally publishedYes
EventWorld Conference on Information Systems and Technologies, WorldCIST 2016 - Recife, Brazil
Duration: 22 Mar 201624 Mar 2016

Publication series

NameAdvances in Intelligent Systems and Computing
Volume444
ISSN (Print)2194-5357

Conference

ConferenceWorld Conference on Information Systems and Technologies, WorldCIST 2016
Country/TerritoryBrazil
CityRecife
Period22/03/1624/03/16

Keywords

  • Email answering
  • Email categorization
  • Text-pattern matching

Fingerprint

Dive into the research topics of 'Email answering by matching question and context-specific text patterns: Performance and error analysis'. Together they form a unique fingerprint.

Cite this