8179

Continuous Partial Order Planning for Multichannel Document Analysis: A Process-Driven Approach

Kristin Stamm, Marcus Liwicki, Andreas Dengel

Proceedings ICDAR 2013 International Conference on Document Analysis and Recognition (ICDAR), Washington D.C. USA , Pages: 626-630 , IEEE , 2013
Abstract—With the rise of email communication, enterprises strive to manage incoming documents from all input channels for achieving customer satisfaction. Their overall goal is to reduce request processing time and to increase processing quality. Previously, we proposed the approach of process-driven document analysis (DA) using the concepts of Attentive Tasks (ATs) and the Specialist Board (SB). The ATs formalize information expectations of the processes toward an incoming document, whereas the SB describes all available DA methods. Here, we propose to apply continuous partial order planning (CPOP) from machine learning for guiding DA with the goal optimal extraction accuracy and runtime. To our knowledge, this approach provides a novel method to integrate knowledge management with DA, in particular for processes. Since planning has not been applied to this field yet, we explore learning the suitability function (SF) and the adaptation of the DA plan. For SF optimization we propose: (1) Suitability measures of runtime, accuracy, and their combination and (2) offline, online, as well as off- + online suitability learning. For planning adaptation strategies we examine: (1) one-time goal setting, (2) continuous current state, and (3) continuous goal adaptation. First evaluations indicate the applicability of the approach and preferences for calibration.

Show BibTex:

@inproceedings {
       abstract = {Abstract—With the rise of email communication, enterprises
strive to manage incoming documents from all input channels
for achieving customer satisfaction. Their overall goal is to
reduce request processing time and to increase processing quality.
Previously, we proposed the approach of process-driven document
analysis (DA) using the concepts of Attentive Tasks (ATs)
and the Specialist Board (SB). The ATs formalize information
expectations of the processes toward an incoming document,
whereas the SB describes all available DA methods. Here, we
propose to apply continuous partial order planning (CPOP) from
machine learning for guiding DA with the goal optimal extraction
accuracy and runtime. To our knowledge, this approach provides
a novel method to integrate knowledge management with DA, in
particular for processes. Since planning has not been applied
to this field yet, we explore learning the suitability function
(SF) and the adaptation of the DA plan. For SF optimization
we propose: (1) Suitability measures of runtime, accuracy, and
their combination and (2) offline, online, as well as off- +
online suitability learning. For planning adaptation strategies we
examine: (1) one-time goal setting, (2) continuous current state,
and (3) continuous goal adaptation. First evaluations indicate the
applicability of the approach and preferences for calibration.},
       number = {}, 
       month = {8}, 
       year = {2013}, 
       title = {Continuous Partial Order Planning for Multichannel Document Analysis: A Process-Driven Approach}, 
       journal = {}, 
       volume = {}, 
       pages = {626-630}, 
       publisher = {IEEE}, 
       author = {Kristin Stamm, Marcus Liwicki, Andreas Dengel}, 
       keywords = {Document Analysis, Information Extraction, Process Management},
       url = {http://www.dfki.de/web/forschung/publikationen/renameFileForDownload?filename=Planning_v04.pdf&file_id=uploads_2687}
}