Section: New Results

Care Pathway Analysis with ASP Sequence Mining

Participants : Ahmed Samet, Benjamin Négrevergne, Thomas Guyet.

This line of work aims at applying our ASP encoding for sequential pattern mining to care pathway analysis (see section 4.3 for applicative objectives). This year, we proposed an approach of meaningful rare sequential pattern mining based on the declarative programming paradigm of Answer Set Programming (ASP). The setting of rare sequential pattern mining is introduced. To cope with the huge amount of meaningless rare patterns, our ASP approach provides an easy manner to encode expert constraints on expected patterns. Encodings are presented and quantitatively compared to a procedural baseline. An application on care pathways analysis illustrates the qualitative interest of expert constraints encoding.

This work has been submitted to the PAKDD 2017 conference.