======= Review 1 =======

*** Strengths: What are the major reasons to accept the paper? [Be brief.]

Interesting topic; Relevant to the WoWMoM community; It will generate discussion among the audience; Well-written; Solid technically.

*** Weaknesses: What are the major reasons NOT to accept the paper? [Be brief.]

Cross-validation of the prediction; The performance analysis can be extended to have more conclusive results; 

Improvement in the presentation (e.g., future work plans, more detailed description of the Figs.)

*** Detailed Comments: Please provide detailed comments that will help the TPC assess the paper and help provide feedback to the authors.

Did you perform a (ten fold) cross validation for the prediction?

How things would change if the data are more noisy?

Have you tried random forests or SVM to see if the performance can be improved?

Did you perform a user-centric analysis/prediction? I would be curious if your prediction accuracy is consistent across users.

Indicate:
i) the (hyper)parameters in running the decision trees. 
ii) the baseline prediction (prediction score in the case of random selection)

Discuss the challenges when you include users in different context/mobility-usage patterns.

Editorial/presentation: 
-- Early on indicate which is the ground truth of your data (GPS)

-- In the Conclusion, indicate more clearly and precisely the key findings, as well as the future work plans (e.g., perhaps repeating the study with a larger set of users, more heterogeneous population to include more diverse context)

-- Improve the caption of the figures to more clearly describe the main observations/trends (e.g.,  Fig 3 is hard to read )

-- Increase the font size of Figures

-- consistent use of abbrevation of seconds (e.g., s or sec)

*** Recommendation: Your overall rating.
Accept (a good paper with minor issues, fair novelty and relevance) (3)

======= Review 2 =======

*** Strengths: What are the major reasons to accept the paper? [Be brief.]

This paper studies an interesting problem on user mobility data that is obtained from mobile phones. There are cases where the mobile phone location data suggest mobility of the user, which in reality wasn't the case. 

The aim of this research is to detect such cases and discount them for producing the mobility pattern of a user. 

The authors use a unique dataset of spatiotemporal individual trajectories that captures both the user and network operator perspectives in mobile phone
location data and investigate the oscillation phenomenon.

*** Weaknesses: What are the major reasons NOT to accept the paper? [Be brief.]

There are no such weaknesses in this paper. 
The reviewer could hardly figure out the differences among the Figures 
4(a), 4(b),4(d), and 4(d).

*** Detailed Comments: Please provide detailed comments that will help the TPC assess the paper and help provide feedback to the authors.

The paper is well-written and easy to read except understanding the interpretations of Figs 3(b) and 3(c).

*** Recommendation: Your overall rating.
Accept (a good paper with minor issues, fair novelty and relevance) (3)