[dcchairs2016] Review of UbiComp/ISWC 2016 Doctoral School submission 115

[dcchairs2016 at ubicomp.org]

Paper  115 - Are You (Not) Entertained? Estimating the state of a crowd in an event using wearable sensors
Reviewer 1 - Max Muhlhauser

Overall rating:  5  (scale is 1..6; 6 is best)

Pre-PC discussion (hidden from author)

   Potential to accept (will be discussed at PC meeting) 

Confidence

   Very confident - I am knowledgeable in the area 

Contribution to UbiComp

   The paper addresses the analysis of crowd behavior aiming at deriving the
   user experience of crowd members. The use of wearables for gathering
   sensor data from the crowd members as a starting point makes the paper
   sufficiently pertinent to UbiComp.
 

Overall Rating

   5  (Probably accept: I would argue for accepting this paper.)

R&R Suitability (hidden from author)

   High potential for significant improvement in 5 weeks 

The Review

   The aim of the PhD research can be briefly summarized as mastering the
   following steps for "crowd events" (speed dating, networking events,
   ...):
   1.	Measurement of sensor data via wearables carried by event participants
   2.	Detection of individual actions
   3.	Detection of interactions
   4.	Inferring of user experience
   Obviously, the work has a strong slant towards machine learning (ML). 
   State of the art reflection is decent, the “methodology and key
   ideas” section is pleasant to read and well structured. The section on
   “conducted and planned research” is also rather elaborate and shows
   that the advancement of the work is about right for the doctoral colloq
   (far enough advanced, still far enough from the defense). 
   A point of critique regards step 4 above: inferring UX is a quite
   ambitious goal, given the complex and in part hedonic nature of UX. The
   paper is not yet convincing w.r.t. the concrete plans here and does not
   create much confidence yet that this step can be achieved.
   If related paper 121 gets accepted, the similarities and differences
   w.r.t. ML approaches should be highlighted at the presentations. 

 

Confidential Comments (Optional) (hidden from author)

 


To see the review, go to https://precisionconference.com/~ubicomp?goto=ubicomp16c